Test stuff

## Motivation

GPU timeouts often when prompt size > profile_step_size. It also happens
for seemingly random models.

## Changes

Add mx.depends for cache on the logits.
All gather at the model level rather than the layer level, reducing the
amount of data sent.

## Why It Works

mlx_lm's prefill loop only evaluates cache state, not logits.
When prompt > prefill_step_size, the all_gather is never evaluated,
causing GPU timeout.

## Test Plan

### Manual Testing
<!-- Hardware: (e.g., MacBook Pro M1 Max 32GB, Mac Mini M2 16GB,
connected via Thunderbolt 4) -->
<!-- What you did: -->
<!-- - -->

### Automated Testing
Added failing test cases and then resolved them.

.github/workflows/build-app.yml

@@ -1,6 +1,18 @@
 name: Build EXO macOS DMG
 
+# Release workflow:
+# 1. Create a draft GitHub Release with the tag name (e.g. v1.0.0) and write release notes in markdown
+# 2. Push the tag: git tag v1.0.0 && git push origin v1.0.0
+# 3. This workflow builds, signs, and notarizes the DMG
+# 4. Release notes are embedded in appcast.xml for Sparkle (rendered as markdown)
+# 5. DMG and appcast.xml are uploaded to S3
+# 6. The draft GitHub Release is published with the DMG attached
+#
+# For alpha releases (e.g. v1.0.0-alpha.1): draft release and notes are optional.
+# If no draft exists, a release is auto-created with generated notes.
+
 on:
+  workflow_dispatch:
   push:
     tags:
       - "v*"
@@ -10,8 +22,10 @@ on:
 jobs:
   build-macos-app:
     runs-on: "macos-26"
+    permissions:
+      contents: write
     env:
-      SPARKLE_VERSION: 2.8.1
+      SPARKLE_VERSION: 2.9.0-beta.1
       SPARKLE_DOWNLOAD_PREFIX: ${{ secrets.SPARKLE_DOWNLOAD_PREFIX }}
       SPARKLE_FEED_URL: ${{ secrets.SPARKLE_FEED_URL }}
       SPARKLE_ED25519_PUBLIC: ${{ secrets.SPARKLE_ED25519_PUBLIC }}
@@ -35,7 +49,7 @@ jobs:
 
       - name: Derive release version from tag
         run: |
-          if [[ "$GITHUB_REF_NAME" == "test-app" ]]; then
+          if [[ "$GITHUB_REF_NAME" == "test-app" || "${{ github.event_name }}" == "workflow_dispatch" ]]; then
             VERSION="0.0.0-alpha.0"
             echo "IS_ALPHA=true" >> $GITHUB_ENV
           else
@@ -48,6 +62,32 @@ jobs:
           fi
           echo "RELEASE_VERSION=$VERSION" >> $GITHUB_ENV
 
+      - name: Compute build version from semver
+        run: |
+          VERSION="$RELEASE_VERSION"
+          # Extract major.minor.patch (strip prerelease suffix)
+          BASE_VERSION="${VERSION%%-*}"
+          MAJOR=$(echo "$BASE_VERSION" | cut -d. -f1)
+          MINOR=$(echo "$BASE_VERSION" | cut -d. -f2)
+          PATCH=$(echo "$BASE_VERSION" | cut -d. -f3)
+
+          # Extract prerelease number (e.g., "alpha.2" -> 2, or 999 for releases)
+          if [[ "$VERSION" == *-* ]]; then
+            PRERELEASE_PART="${VERSION#*-}"
+            PRERELEASE_NUM="${PRERELEASE_PART##*.}"
+            # Default to 0 if not a number
+            if ! [[ "$PRERELEASE_NUM" =~ ^[0-9]+$ ]]; then
+              PRERELEASE_NUM=0
+            fi
+          else
+            PRERELEASE_NUM=999
+          fi
+
+          # Compute: PRERELEASE + (1000 * PATCH) + (1_000_000 * MINOR) + (1_000_000_000 * MAJOR)
+          BUILD_VERSION=$((PRERELEASE_NUM + 1000 * PATCH + 1000000 * MINOR + 1000000000 * MAJOR))
+          echo "EXO_BUILD_VERSION=$BUILD_VERSION" >> $GITHUB_ENV
+          echo "Computed build version: $BUILD_VERSION from $VERSION"
+
       - name: Ensure tag commit is on main
         if: github.ref_type == 'tag'
         run: |
@@ -60,6 +100,52 @@ jobs:
             exit 1
           fi
 
+      - name: Fetch and validate release notes
+        if: github.ref_type == 'tag'
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          # Find draft release by name using gh release list (more reliable with default token)
+          echo "Looking for draft release named '$GITHUB_REF_NAME'..."
+          DRAFT_EXISTS=$(gh release list --json name,isDraft --jq ".[] | select(.isDraft == true) | select(.name == \"$GITHUB_REF_NAME\") | .name" 2>/dev/null || echo "")
+
+          if [[ -z "$DRAFT_EXISTS" ]]; then
+            if [[ "$IS_ALPHA" == "true" ]]; then
+              echo "No draft release found for alpha tag $GITHUB_REF_NAME (optional for alphas)"
+              echo "HAS_RELEASE_NOTES=false" >> $GITHUB_ENV
+              exit 0
+            fi
+            echo "ERROR: No draft release found for tag $GITHUB_REF_NAME"
+            echo "Please create a draft release with release notes before pushing the tag."
+            exit 1
+          fi
+
+          # Fetch full release details via API to get body and ID
+          echo "Found draft release, fetching details..."
+          RELEASE_JSON=$(gh api repos/${{ github.repository }}/releases --jq ".[] | select(.draft == true) | select(.name == \"$GITHUB_REF_NAME\")" 2>/dev/null || echo "")
+
+          # Extract release notes
+          NOTES=$(echo "$RELEASE_JSON" | jq -r '.body // ""')
+          if [[ -z "$NOTES" || "$NOTES" == "null" ]]; then
+            if [[ "$IS_ALPHA" == "true" ]]; then
+              echo "Draft release has no notes (optional for alphas)"
+              echo "HAS_RELEASE_NOTES=false" >> $GITHUB_ENV
+              exit 0
+            fi
+            echo "ERROR: Draft release exists but has no release notes"
+            echo "Please add release notes to the draft release before pushing the tag."
+            exit 1
+          fi
+
+          # Save release ID for later publishing
+          RELEASE_ID=$(echo "$RELEASE_JSON" | jq -r '.id')
+          echo "DRAFT_RELEASE_ID=$RELEASE_ID" >> $GITHUB_ENV
+          echo "HAS_RELEASE_NOTES=true" >> $GITHUB_ENV
+
+          echo "Found draft release (ID: $RELEASE_ID), saving release notes..."
+          echo "$NOTES" > /tmp/release_notes.md
+          echo "RELEASE_NOTES_FILE=/tmp/release_notes.md" >> $GITHUB_ENV
+
       # ============================================================
       # Install dependencies
       # ============================================================
@@ -86,11 +172,22 @@ jobs:
           uv python install
           uv sync --locked
 
+      - name: Install Nix
+        uses: cachix/install-nix-action@v31
+        with:
+          nix_path: nixpkgs=channel:nixos-unstable
+
+      - name: Configure Cachix
+        uses: cachix/cachix-action@v14
+        with:
+          name: exo
+          authToken: "${{ secrets.CACHIX_AUTH_TOKEN }}"
+
       - name: Build dashboard
         run: |
-          cd dashboard
-          npm ci
-          npm run build
+          DASHBOARD_OUT=$(nix build .#dashboard --print-build-logs --no-link --print-out-paths)
+          mkdir -p dashboard/build
+          cp -r "$DASHBOARD_OUT"/* dashboard/build/
 
       - name: Install Sparkle CLI
         run: |
@@ -163,7 +260,7 @@ jobs:
             -configuration Release \
             -derivedDataPath build \
             MARKETING_VERSION="$RELEASE_VERSION" \
-            CURRENT_PROJECT_VERSION="$EXO_BUILD_NUMBER" \
+            CURRENT_PROJECT_VERSION="$EXO_BUILD_VERSION" \
             EXO_BUILD_TAG="$RELEASE_VERSION" \
             EXO_BUILD_COMMIT="$GITHUB_SHA" \
             SPARKLE_FEED_URL="$SPARKLE_FEED_URL" \
@@ -266,6 +363,28 @@ jobs:
             $CHANNEL_FLAG \
             .
 
+      - name: Inject release notes into appcast
+        if: github.ref_type == 'tag' && env.HAS_RELEASE_NOTES == 'true'
+        env:
+          RELEASE_VERSION: ${{ env.RELEASE_VERSION }}
+        run: |
+          # Inject markdown release notes with sparkle:format="markdown" (Sparkle 2.9+)
+          export NOTES=$(cat "$RELEASE_NOTES_FILE")
+
+          # Insert description after the enclosure tag for this version
+          awk '
+            /<enclosure[^>]*>/ && index($0, ENVIRON["RELEASE_VERSION"]) {
+              print
+              print "            <description sparkle:format=\"markdown\"><![CDATA["
+              print ENVIRON["NOTES"]
+              print "            ]]></description>"
+              next
+            }
+            { print }
+          ' output/appcast.xml > output/appcast.xml.tmp && mv output/appcast.xml.tmp output/appcast.xml
+
+          echo "Injected markdown release notes for version $RELEASE_VERSION"
+
       # ============================================================
       # Upload artifacts
       # ============================================================
@@ -296,5 +415,28 @@ jobs:
           aws s3 cp "$DMG_NAME" "s3://${SPARKLE_S3_BUCKET}/${PREFIX}${DMG_NAME}"
           if [[ "$IS_ALPHA" != "true" ]]; then
             aws s3 cp "$DMG_NAME" "s3://${SPARKLE_S3_BUCKET}/${PREFIX}EXO-latest.dmg"
+            aws s3 cp appcast.xml "s3://${SPARKLE_S3_BUCKET}/${PREFIX}appcast.xml" --content-type application/xml --cache-control no-cache
+          fi
+
+      - name: Publish GitHub Release
+        if: github.ref_type == 'tag'
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          DMG_PATH="output/EXO-${RELEASE_VERSION}.dmg"
+
+          if [[ "$HAS_RELEASE_NOTES" == "true" ]]; then
+            # Update the draft release with the tag and upload DMG
+            gh api --method PATCH "repos/${{ github.repository }}/releases/$DRAFT_RELEASE_ID" \
+              -f tag_name="$GITHUB_REF_NAME" \
+              -F draft=false
+            gh release upload "$GITHUB_REF_NAME" "$DMG_PATH" --clobber
+            echo "Published release $GITHUB_REF_NAME with DMG attached"
+          else
+            # Alpha without draft release - create one with auto-generated notes
+            gh release create "$GITHUB_REF_NAME" "$DMG_PATH" \
+              --title "$GITHUB_REF_NAME" \
+              --generate-notes \
+              --prerelease
+            echo "Created alpha release $GITHUB_REF_NAME with auto-generated notes"
           fi
-          aws s3 cp appcast.xml "s3://${SPARKLE_S3_BUCKET}/${PREFIX}appcast.xml" --content-type application/xml --cache-control no-cache

.github/workflows/pipeline.yml

@@ -20,6 +20,12 @@ jobs:
         with:
           nix_path: nixpkgs=channel:nixos-unstable
 
+      - uses: cachix/cachix-action@v14
+        name: Configure Cachix
+        with:
+          name: exo
+          authToken: "${{ secrets.CACHIX_AUTH_TOKEN }}"
+
       - name: Configure git user
         run: |
           git config --local user.email "github-actions@users.noreply.github.com"
@@ -88,9 +94,19 @@ jobs:
 
       - uses: ./.github/actions/typecheck
 
-  nix-flake-check:
-    name: Check Nix flake
-    runs-on: ubuntu-latest
+  nix:
+    name: Build and check (${{ matrix.system }})
+    runs-on: ${{ matrix.runner }}
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - runner: macos-26
+            system: aarch64-darwin
+          - runner: ubuntu-latest
+            system: x86_64-linux
+          - runner: ubuntu-24.04-arm
+            system: aarch64-linux
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
@@ -101,83 +117,20 @@ jobs:
         with:
           nix_path: nixpkgs=channel:nixos-unstable
 
-      - name: Run nix flake check
-        run: |
-          nix flake check
-        shell: bash
+      - uses: cachix/cachix-action@v14
+        name: Configure Cachix
+        with:
+          name: exo
+          authToken: "${{ secrets.CACHIX_AUTH_TOKEN }}"
 
-#  ci:
-#    needs: typecheck
-#    runs-on: ubuntu-latest
-#    permissions:
-#      contents: read
-#    env:
-#      GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-#    steps:
-#      - name: Checkout repository
-#        uses: actions/checkout@v4
-#        with:
-#          fetch-depth: 0
-#          token: ${{ secrets.GITHUB_TOKEN }}
-#          lfs: true
-#
-#      - name: Configure git user
-#        run: |
-#          git config --local user.email "github-actions@users.noreply.github.com"
-#          git config --local user.name  "github-actions bot"
-#        shell: bash
-#
-#      - name: Pull LFS files
-#        run: |
-#          echo "Pulling Git LFS files..."
-#          git lfs pull
-#        shell: bash
-#
-#      - name: Setup EXO_HOME and API_PORT
-#        run: |
-#          EXO_HOME=$(mktemp -d -t exo-ci-XXXXXXXX)
-#          # Generate random port (macOS compatible method)
-#          API_PORT=$((49152 + RANDOM % (65535 - 49152 + 1)))
-#          echo "EXO_HOME=$EXO_HOME" >> $GITHUB_ENV
-#          echo "API_PORT=$API_PORT" >> $GITHUB_ENV
-#          echo "Created EXO_HOME: $EXO_HOME"
-#          echo "Generated API_PORT: $API_PORT"
-#        shell: bash
-#
-#      - name: Setup Nix Environment
-#        run: |
-#          echo "Checking for nix installation..."
-#          
-#          # Check if nix binary exists directly
-#          if [ -f /nix/var/nix/profiles/default/bin/nix ]; then
-#            echo "Found nix binary at /nix/var/nix/profiles/default/bin/nix"
-#            export PATH="/nix/var/nix/profiles/default/bin:$PATH"
-#            echo "PATH=$PATH" >> $GITHUB_ENV
-#            nix --version
-#          elif [ -f /nix/var/nix/profiles/default/etc/profile.d/nix-daemon.sh ]; then
-#            echo "Found nix profile script, sourcing..."
-#            source /nix/var/nix/profiles/default/etc/profile.d/nix-daemon.sh
-#            nix --version
-#          elif command -v nix >/dev/null 2>&1; then
-#            echo "Nix already in PATH"
-#            nix --version
-#          else
-#            echo "Nix not found. Debugging info:"
-#            echo "Contents of /nix/var/nix/profiles/default/:"
-#            ls -la /nix/var/nix/profiles/default/ 2>/dev/null || echo "Directory not found"
-#            echo "Contents of /nix/var/nix/profiles/default/bin/:"
-#            ls -la /nix/var/nix/profiles/default/bin/ 2>/dev/null || echo "Directory not found"
-#            exit 1
-#          fi
-#        shell: bash
-#
-#      - uses: ./.github/actions/lint-check
-#
-#      - uses: ./.github/actions/unit-test
-#
-#      - name: Cleanup EXO_HOME
-#        run: |
-#          echo "Cleaning up EXO_HOME: $EXO_HOME"
-#          rm -rf "$EXO_HOME"
-#        shell: bash
-#        if: always()
+      - name: Build all Nix outputs
+        run: |
+          nix flake show --json | jq -r '
+            [
+              (.packages."${{ matrix.system }}" // {} | keys[] | ".#packages.${{ matrix.system }}.\(.)"),
+              (.devShells."${{ matrix.system }}" // {} | keys[] | ".#devShells.${{ matrix.system }}.\(.)")
+            ] | .[]
+          ' | xargs nix build
+
+      - name: Run nix flake check
+        run: nix flake check

.gitignore

@@ -16,6 +16,7 @@ digest.txt
 *.xcuserdatad/
 **/.DS_Store
 app/EXO/build/
+dist/
 
 
 # rust

.mlx_typings/mlx_lm/models/deepseek_v3.pyi

@@ -0,0 +1,156 @@
+"""Type stubs for mlx_lm.models.deepseek_v3"""
+
+from dataclasses import dataclass
+from typing import Any, Dict, Optional
+
+import mlx.core as mx
+import mlx.nn as nn
+
+from .base import BaseModelArgs
+from .switch_layers import SwitchGLU
+
+@dataclass
+class ModelArgs(BaseModelArgs):
+    model_type: str
+    vocab_size: int
+    hidden_size: int
+    intermediate_size: int
+    moe_intermediate_size: int
+    num_hidden_layers: int
+    num_attention_heads: int
+    num_key_value_heads: int
+    n_shared_experts: Optional[int]
+    n_routed_experts: Optional[int]
+    routed_scaling_factor: float
+    kv_lora_rank: int
+    q_lora_rank: Optional[int]
+    qk_rope_head_dim: int
+    v_head_dim: int
+    qk_nope_head_dim: int
+    topk_method: str
+    scoring_func: str
+    norm_topk_prob: bool
+    n_group: int
+    topk_group: int
+    num_experts_per_tok: int
+    moe_layer_freq: int
+    first_k_dense_replace: int
+    max_position_embeddings: int
+    rms_norm_eps: float
+    rope_theta: float
+    rope_scaling: Optional[Dict[str, Any]]
+    attention_bias: bool
+
+class DeepseekV3Attention(nn.Module):
+    config: ModelArgs
+    hidden_size: int
+    num_heads: int
+    max_position_embeddings: int
+    rope_theta: float
+    q_lora_rank: Optional[int]
+    qk_rope_head_dim: int
+    kv_lora_rank: int
+    v_head_dim: int
+    qk_nope_head_dim: int
+    q_head_dim: int
+    scale: float
+    q_proj: nn.Linear
+    q_a_proj: nn.Linear
+    q_a_layernorm: nn.RMSNorm
+    q_b_proj: nn.Linear
+    kv_a_proj_with_mqa: nn.Linear
+    kv_a_layernorm: nn.RMSNorm
+    kv_b_proj: nn.Linear
+    o_proj: nn.Linear
+    rope: Any
+
+    def __init__(self, config: ModelArgs) -> None: ...
+    def __call__(
+        self,
+        x: mx.array,
+        mask: Optional[mx.array] = None,
+        cache: Optional[Any] = None,
+    ) -> mx.array: ...
+
+class DeepseekV3MLP(nn.Module):
+    config: ModelArgs
+    hidden_size: int
+    intermediate_size: int
+    gate_proj: nn.Linear
+    up_proj: nn.Linear
+    down_proj: nn.Linear
+
+    def __init__(
+        self,
+        config: ModelArgs,
+        hidden_size: Optional[int] = None,
+        intermediate_size: Optional[int] = None,
+    ) -> None: ...
+    def __call__(self, x: mx.array) -> mx.array: ...
+
+class MoEGate(nn.Module):
+    config: ModelArgs
+    top_k: int
+    norm_topk_prob: bool
+    n_routed_experts: Optional[int]
+    routed_scaling_factor: float
+    n_group: int
+    topk_group: int
+    weight: mx.array
+    e_score_correction_bias: mx.array
+
+    def __init__(self, config: ModelArgs) -> None: ...
+    def __call__(self, x: mx.array) -> tuple[mx.array, mx.array]: ...
+
+class DeepseekV3MoE(nn.Module):
+    config: ModelArgs
+    num_experts_per_tok: int
+    switch_mlp: SwitchGLU
+    gate: MoEGate
+    shared_experts: DeepseekV3MLP
+    sharding_group: Optional[mx.distributed.Group]
+
+    def __init__(self, config: ModelArgs) -> None: ...
+    def __call__(self, x: mx.array) -> mx.array: ...
+
+class DeepseekV3DecoderLayer(nn.Module):
+    self_attn: DeepseekV3Attention
+    mlp: DeepseekV3MLP | DeepseekV3MoE
+    input_layernorm: nn.RMSNorm
+    post_attention_layernorm: nn.RMSNorm
+
+    def __init__(self, config: ModelArgs, layer_idx: int) -> None: ...
+    def __call__(
+        self,
+        x: mx.array,
+        mask: Optional[mx.array] = None,
+        cache: Optional[Any] = None,
+    ) -> mx.array: ...
+
+class DeepseekV3Model(nn.Module):
+    vocab_size: int
+    embed_tokens: nn.Embedding
+    layers: list[DeepseekV3DecoderLayer]
+    norm: nn.RMSNorm
+
+    def __init__(self, config: ModelArgs) -> None: ...
+    def __call__(
+        self,
+        x: mx.array,
+        cache: Optional[Any] = None,
+    ) -> mx.array: ...
+
+class Model(nn.Module):
+    model_type: str
+    model: DeepseekV3Model
+    lm_head: nn.Linear
+
+    def __init__(self, config: ModelArgs) -> None: ...
+    def __call__(
+        self,
+        inputs: mx.array,
+        cache: Optional[Any] = None,
+    ) -> mx.array: ...
+    def sanitize(self, weights: dict[str, Any]) -> dict[str, Any]: ...
+    @property
+    def layers(self) -> list[DeepseekV3DecoderLayer]: ...

.mlx_typings/mlx_lm/models/switch_layers.pyi

@@ -57,6 +57,11 @@ class SwiGLU(nn.Module):
     def __call__(self, x, gate): ...
 
 class SwitchGLU(nn.Module):
+    gate_proj: SwitchLinear
+    up_proj: SwitchLinear
+    down_proj: SwitchLinear
+    activation: SwiGLU
+
     def __init__(
         self,
         input_dims: int,

.mlx_typings/mlx_lm/tokenizer_utils.pyi

@@ -4,6 +4,7 @@ This type stub file was generated by pyright.
 
 from functools import partial
 from pathlib import Path
+from typing import Any
 
 from transformers import PreTrainedTokenizerFast
 
@@ -103,37 +104,55 @@ class TokenizerWrapper:
     Accessing any attribute other than the ``detokenizer`` is forwarded to the
     huggingface tokenizer.
     """
-    def __init__(self, tokenizer, detokenizer_class=..., eos_token_ids=...) -> None: ...
-    def add_eos_token(self, token: str):  # -> None:
-        ...
-    @property
-    def has_thinking(self):  # -> bool:
-        ...
-    @property
-    def think_start(self):  # -> str | None:
-        ...
-    @property
-    def think_end(self):  # -> str | None:
-        ...
-    @property
-    def has_tool_calling(self):  # -> bool:
-        ...
-    @property
-    def tool_call_start(self):  # -> str | None:
-        ...
-    @property
-    def tool_call_end(self):  # -> str | None:
-        ...
-    @property
-    def detokenizer(self):  # -> NaiveStreamingDetokenizer:
-        """
-        Get a stateful streaming detokenizer.
-        """
 
-    def __getattr__(self, attr):  # -> set[Any] | Any:
-        ...
-    def __setattr__(self, attr, value):  # -> None:
-        ...
+    _tokenizer: PreTrainedTokenizerFast
+    eos_token_id: int | None
+    eos_token: str | None
+    bos_token_id: int | None
+    bos_token: str | None
+    vocab_size: int
+    all_special_tokens: list[str]
+
+    def __init__(
+        self,
+        tokenizer: Any,
+        detokenizer_class: Any = ...,
+        eos_token_ids: list[int] | None = ...,
+        chat_template: Any = ...,
+        tool_parser: Any = ...,
+        tool_call_start: str | None = ...,
+        tool_call_end: str | None = ...,
+    ) -> None: ...
+    def encode(self, text: str, **kwargs: Any) -> list[int]: ...
+    def decode(self, token_ids: list[int], **kwargs: Any) -> str: ...
+    def apply_chat_template(
+        self,
+        messages: list[dict[str, Any]],
+        tokenize: bool = False,
+        add_generation_prompt: bool = False,
+        tools: Any = None,
+        **kwargs: Any,
+    ) -> str: ...
+    def get_vocab(self) -> dict[str, int]: ...
+    def add_eos_token(self, token: str) -> None: ...
+    @property
+    def has_thinking(self) -> bool: ...
+    @property
+    def think_start(self) -> str | None: ...
+    @property
+    def think_end(self) -> str | None: ...
+    @property
+    def has_tool_calling(self) -> bool: ...
+    @property
+    def tool_call_start(self) -> str | None: ...
+    @property
+    def tool_call_end(self) -> str | None: ...
+    @property
+    def detokenizer(self) -> NaiveStreamingDetokenizer:
+        """Get a stateful streaming detokenizer."""
+
+    def __getattr__(self, attr: str) -> Any: ...
+    def __setattr__(self, attr: str, value: Any) -> None: ...
 
 class NewlineTokenizer(PreTrainedTokenizerFast):
     """A tokenizer that replaces newlines with <n> and <n> with new line."""
@@ -146,18 +165,11 @@ class NewlineTokenizer(PreTrainedTokenizerFast):
     def batch_decode(self, *args, **kwargs):  # -> list[str]:
         ...
 
-def load_tokenizer(
+def load(
     model_path: Path,
-    tokenizer_config_extra=...,
-    return_tokenizer=...,
-    eos_token_ids=...,
-) -> (
-    TokenizerWrapper
-    | type[SPMStreamingDetokenizer]
-    | partial[SPMStreamingDetokenizer]
-    | type[BPEStreamingDetokenizer]
-    | type[NaiveStreamingDetokenizer]
-):
+    tokenizer_config_extra: dict[str, Any] | None = None,
+    eos_token_ids: list[int] | int | None = None,
+) -> TokenizerWrapper:
     """Load a huggingface tokenizer and try to infer the type of streaming
     detokenizer to use.
 
@@ -165,4 +177,7 @@ def load_tokenizer(
     a Hugging Face repo ID.
     """
 
-def no_bos_or_eos(sequence: list, bos: int, eos: int) -> list: ...
+# Alias for backward compatibility
+load_tokenizer = load
+
+def no_bos_or_eos(sequence: list[int], bos: int, eos: int) -> list[int]: ...

AGENTS.md

@@ -0,0 +1,121 @@
+# AGENTS.md
+
+This file provides guidance to AI coding agents when working with code in this repository.
+
+## Project Overview
+
+exo is a distributed AI inference system that connects multiple devices into a cluster. It enables running large language models across multiple machines using MLX as the inference backend and libp2p for peer-to-peer networking.
+
+## Build & Run Commands
+
+```bash
+# Build the dashboard (required before running exo)
+cd dashboard && npm install && npm run build && cd ..
+
+# Run exo (starts both master and worker with API at http://localhost:52415)
+uv run exo
+
+# Run with verbose logging
+uv run exo -v   # or -vv for more verbose
+
+# Run tests (excludes slow tests by default)
+uv run pytest
+
+# Run all tests including slow tests
+uv run pytest -m ""
+
+# Run a specific test file
+uv run pytest src/exo/shared/tests/test_election.py
+
+# Run a specific test function
+uv run pytest src/exo/shared/tests/test_election.py::test_function_name
+
+# Type checking (strict mode)
+uv run basedpyright
+
+# Linting
+uv run ruff check
+
+# Format code (using nix)
+nix fmt
+```
+
+## Pre-Commit Checks (REQUIRED)
+
+**IMPORTANT: Always run these checks before committing code. CI will fail if these don't pass.**
+
+```bash
+# 1. Type checking - MUST pass with 0 errors
+uv run basedpyright
+
+# 2. Linting - MUST pass
+uv run ruff check
+
+# 3. Formatting - MUST be applied
+nix fmt
+
+# 4. Tests - MUST pass
+uv run pytest
+```
+
+Run all checks in sequence:
+```bash
+uv run basedpyright && uv run ruff check && nix fmt && uv run pytest
+```
+
+If `nix fmt` changes any files, stage them before committing. The CI runs `nix flake check` which verifies formatting, linting, and runs Rust tests.
+
+## Architecture
+
+### Node Composition
+A single exo `Node` (src/exo/main.py) runs multiple components:
+- **Router**: libp2p-based pub/sub messaging via Rust bindings (exo_pyo3_bindings)
+- **Worker**: Handles inference tasks, downloads models, manages runner processes
+- **Master**: Coordinates cluster state, places model instances across nodes
+- **Election**: Bully algorithm for master election
+- **API**: FastAPI server for OpenAI-compatible chat completions
+
+### Message Flow
+Components communicate via typed pub/sub topics (src/exo/routing/topics.py):
+- `GLOBAL_EVENTS`: Master broadcasts indexed events to all workers
+- `LOCAL_EVENTS`: Workers send events to master for indexing
+- `COMMANDS`: Workers/API send commands to master
+- `ELECTION_MESSAGES`: Election protocol messages
+- `CONNECTION_MESSAGES`: libp2p connection updates
+
+### Event Sourcing
+The system uses event sourcing for state management:
+- `State` (src/exo/shared/types/state.py): Immutable state object
+- `apply()` (src/exo/shared/apply.py): Pure function that applies events to state
+- Master indexes events and broadcasts; workers apply indexed events
+
+### Key Type Hierarchy
+- `src/exo/shared/types/`: Pydantic models for all shared types
+  - `events.py`: Event types (discriminated union)
+  - `commands.py`: Command types
+  - `tasks.py`: Task types for worker execution
+  - `state.py`: Cluster state model
+
+### Rust Components
+Rust code in `rust/` provides:
+- `networking`: libp2p networking (gossipsub, peer discovery)
+- `exo_pyo3_bindings`: PyO3 bindings exposing Rust to Python
+- `system_custodian`: System-level operations
+
+### Dashboard
+Svelte 5 + TypeScript frontend in `dashboard/`. Build output goes to `dashboard/build/` and is served by the API.
+
+## Code Style Requirements
+
+From .cursorrules:
+- Strict, exhaustive typing - never bypass the type-checker
+- Use `Literal[...]` for enum-like sets, `typing.NewType` for primitives
+- Pydantic models with `frozen=True` and `strict=True`
+- Pure functions with injectable effect handlers for side-effects
+- Descriptive names - no abbreviations or 3-letter acronyms
+- Catch exceptions only where you can handle them meaningfully
+- Use `@final` and immutability wherever applicable
+
+## Testing
+
+Tests use pytest-asyncio with `asyncio_mode = "auto"`. Tests are in `tests/` subdirectories alongside the code they test. The `EXO_TESTS=1` env var is set during tests.

CLAUDE.md

@@ -0,0 +1 @@
+AGENTS.md

Cargo.lock

@@ -4340,25 +4340,6 @@ dependencies = [
  "libc",
 ]
 
-[[package]]
-name = "system_custodian"
-version = "0.0.1"
-dependencies = [
- "delegate",
- "derive_more",
- "either",
- "extend",
- "futures",
- "futures-timer",
- "impl-trait-for-tuples",
- "keccak-const",
- "log",
- "thiserror 2.0.17",
- "tokio",
- "tracing-subscriber",
- "util",
-]
-
 [[package]]
 name = "tagptr"
 version = "0.2.0"

Cargo.toml

@@ -3,7 +3,6 @@ resolver = "3"
 members = [
     "rust/networking",
     "rust/exo_pyo3_bindings",
-    "rust/system_custodian",
     "rust/util",
 ]
 
@@ -25,7 +24,6 @@ opt-level = 3
 [workspace.dependencies]
 ## Crate members as common dependencies
 networking = { path = "rust/networking" }
-system_custodian = { path = "rust/system_custodian" }
 util = { path = "rust/util" }
 
 # Proc-macro authoring tools

MISSED_THINGS.md

@@ -0,0 +1,41 @@
+# Missed things
+[X] Log EXO_LIBP2P_NAMESPACE on start in exo/main.py
+[X] Ordering of warmup was changed, which is wrong. It was changed to rank < n-1, then rank=n-1. It should be rank!=0 then rank=0 (this matches the auto_parallel implementation. NOTE: we use a different convention to mlx-lm, our terminal rank is rank=n-1 whereas mlx-lm is rank=0 hence i can see why this was changed wrongly).
+[X] Downloads keying by model_id not shard_metadata (worker/plan.py, worker/main.py).
+[X] Fetching download status of all models on start
+[X] Deduplication of tasks in plan_step.
+[X] resolve_allow_patterns should just be wildcard now.
+[] no mx_barrier in genreate.py mlx_generate at the end.
+[] cache assertion not needed in auto_parallel.py PipelineLastLayer.
+[] GPTOSS support dropped in auto_parallel.py.
+[] sharding changed "all-to-sharded" became _all_to_sharded in auto_parallel.py.
+[] same as above with "sharded-to-all" became _sharded_to_all in auto_parallel.py.
+[] Dropped support for Ministral3Model, DeepseekV32Model, Glm4MoeModel, Qwen3NextModel, GptOssMode in auto_parallel.py.
+[] Dropped prefill/decode code in auto_parallel.py and utils_mlx.py.
+[X] KV_CACHE_BITS should be None to disable quantized KV cache.
+[] Dropped _set_nofile_limit in utils_mlx.py.
+[] We have group optional in load_mlx_items in utils_mlx.py.
+[] Dropped add_missing_chat_templates for GptOss in load_mlx_items in utils_mlx.py.
+[] Dropped model.make_cache in make_kv_cache in utils_mlx.py.
+[X] We put cache limit back in utils_mlx.py.
+[] topology.py remove_node removes the connections after checking if node is is in self._node_id_to_rx_id_map. on beta_1 it checks after, so would remove stale connections I guess?
+[] Missing Glm 4.7 model cards (this isn't ready yet but should be picked up, probably create an issue... the blocker is transforemrs version doesn't support the tokenizer for Glm 4.7. rc-1 does but we can't upgrade as it breaks other things.)
+[] try-except in _command_processor only excepts ValueError. This was silently failing leading to un-debuggable errors (we had a KeyError that was happening ). Changed this to catch Exception instead of ValueError. See exo-v2 89ae38405e0052e3c22405daf094b065878aa873 and fb99fea69b5a39017efc90c5dad0072e677455f0.
+[X] In placement.py, place_instance no longer looks at model_meta.supports_tensor and check if this tensor parallel number of nodes is supported by the model's tensor dimensions.
+[X] In placement.py, place_instanec, we no longer have the special case to exclude DeepSeek v3.1 pipeline parallel (it doesn't work).
+[] logger.warning("You have likely selected ibv for a single node instance; falling back to MlxRing") was changed to debug. That will spam this warning since it happens every time we query instance previews.
+[X] In placement_utils.py, get_mlx_jaccl_coordinators, We no longer prioritise Jaccl Coordinator IP. Now it picks the first one, which is unstable (Jaccl coordinator over TB5 is unstable).
+
+
+
+[X] Downloads keying by model_id not shard_metadata (worker/plan.py, worker/main.py).
+[X] Fetching download status of all models on start
+[X] Deduplication of tasks in plan_step.
+[X] resolve_allow_patterns should just be wildcard now.
+[X] KV_CACHE_BITS should be None to disable quantized KV cache.
+[X] We put cache limit back in utils_mlx.py.
+[X] In placement.py, place_instance no longer looks at model_meta.supports_tensor and check if this tensor parallel number of nodes is supported by the model's tensor dimensions.
+[X] In placement.py, place_instanec, we no longer have the special case to exclude DeepSeek v3.1 pipeline parallel (it doesn't work).
+[X] In placement_utils.py, get_mlx_jaccl_coordinators, We no longer prioritise Jaccl Coordinator IP. Now it picks the first one, which is unstable (Jaccl coordinator over TB5 is unstable).
+
+

README.md

@@ -27,13 +27,22 @@ exo connects all your devices into an AI cluster. Not only does exo enable runni
 - **Tensor Parallelism**: exo supports sharding models, for up to 1.8x speedup on 2 devices and 3.2x speedup on 4 devices.
 - **MLX Support**: exo uses [MLX](https://github.com/ml-explore/mlx) as an inference backend and [MLX distributed](https://ml-explore.github.io/mlx/build/html/usage/distributed.html) for distributed communication.
 
+## Dashboard
+
+exo includes a built-in dashboard for managing your cluster and chatting with models.
+
+<p align="center">
+  <img src="docs/imgs/dashboard-cluster-view.png" alt="exo dashboard - cluster view showing 4 x M3 Ultra Mac Studio with DeepSeek v3.1 and Kimi-K2-Thinking loaded" width="80%" />
+</p>
+<p align="center"><em>4 × 512GB M3 Ultra Mac Studio running DeepSeek v3.1 (8-bit) and Kimi-K2-Thinking (4-bit)</em></p>
+
 ## Benchmarks
 
 <details>
   <summary>Qwen3-235B (8-bit) on 4 × M3 Ultra Mac Studio with Tensor Parallel RDMA</summary>
   <img src="docs/benchmarks/jeffgeerling/mac-studio-cluster-ai-full-1-qwen3-235b.jpeg" alt="Benchmark - Qwen3-235B (8-bit) on 4 × M3 Ultra Mac Studio with Tensor Parallel RDMA" width="80%" />
   <p>
-    <strong>Source:</strong> <a href="https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-studio-rdma-over-thunderbolt-5">Jeff Geerling: 15 TB VRAM on Mac Studio – RDMA over Thunderbolt 5</a>
+    <strong>Source:</strong> <a href="https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-studio-rdma-over-thunderbolt-5">Jeff Geerling: 15 TB VRAM on Mac Studio – RDMA over Thunderbolt 5</a>
   </p>
 </details>
 
@@ -41,7 +50,7 @@ exo connects all your devices into an AI cluster. Not only does exo enable runni
   <summary>DeepSeek v3.1 671B (8-bit) on 4 × M3 Ultra Mac Studio with Tensor Parallel RDMA</summary>
   <img src="docs/benchmarks/jeffgeerling/mac-studio-cluster-ai-full-2-deepseek-3.1-671b.jpeg" alt="Benchmark - DeepSeek v3.1 671B (8-bit) on 4 × M3 Ultra Mac Studio with Tensor Parallel RDMA" width="80%" />
   <p>
-    <strong>Source:</strong> <a href="https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-studio-rdma-over-thunderbolt-5">Jeff Geerling: 15 TB VRAM on Mac Studio – RDMA over Thunderbolt 5</a>
+    <strong>Source:</strong> <a href="https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-studio-rdma-over-thunderbolt-5">Jeff Geerling: 15 TB VRAM on Mac Studio – RDMA over Thunderbolt 5</a>
   </p>
 </details>
 
@@ -49,7 +58,7 @@ exo connects all your devices into an AI cluster. Not only does exo enable runni
   <summary>Kimi K2 Thinking (native 4-bit) on 4 × M3 Ultra Mac Studio with Tensor Parallel RDMA</summary>
   <img src="docs/benchmarks/jeffgeerling/mac-studio-cluster-ai-full-3-kimi-k2-thinking.jpeg" alt="Benchmark - Kimi K2 Thinking (native 4-bit) on 4 × M3 Ultra Mac Studio with Tensor Parallel RDMA" width="80%" />
   <p>
-    <strong>Source:</strong> <a href="https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-studio-rdma-over-thunderbolt-5">Jeff Geerling: 15 TB VRAM on Mac Studio – RDMA over Thunderbolt 5</a>
+    <strong>Source:</strong> <a href="https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-studio-rdma-over-thunderbolt-5">Jeff Geerling: 15 TB VRAM on Mac Studio – RDMA over Thunderbolt 5</a>
   </p>
 </details>
 
@@ -154,6 +163,24 @@ This starts the exo dashboard and API at http://localhost:52415/
 
 **Important note for Linux users:** Currently, exo runs on CPU on Linux. GPU support for Linux platforms is under development. If you'd like to see support for your specific Linux hardware, please [search for existing feature requests](https://github.com/exo-explore/exo/issues) or create a new one.
 
+**Configuration Options:**
+
+- `--no-worker`: Run exo without the worker component. Useful for coordinator-only nodes that handle networking and orchestration but don't execute inference tasks. This is helpful for machines without sufficient GPU resources but with good network connectivity.
+
+  ```bash
+  uv run exo --no-worker
+  ```
+
+**File Locations (Linux):**
+
+exo follows the [XDG Base Directory Specification](https://specifications.freedesktop.org/basedir-spec/basedir-spec-latest.html) on Linux:
+
+- **Configuration files**: `~/.config/exo/` (or `$XDG_CONFIG_HOME/exo/`)
+- **Data files**: `~/.local/share/exo/` (or `$XDG_DATA_HOME/exo/`)
+- **Cache files**: `~/.cache/exo/` (or `$XDG_CACHE_HOME/exo/`)
+
+You can override these locations by setting the corresponding XDG environment variables.
+
 ### macOS App
 
 exo ships a macOS app that runs in the background on your Mac.
@@ -166,6 +193,37 @@ Download the latest build here: [EXO-latest.dmg](https://assets.exolabs.net/EXO-
 
 The app will ask for permission to modify system settings and install a new Network profile. Improvements to this are being worked on.
 
+**Custom Namespace for Cluster Isolation:**
+
+The macOS app includes a custom namespace feature that allows you to isolate your exo cluster from others on the same network. This is configured through the `EXO_LIBP2P_NAMESPACE` setting:
+
+- **Use cases**:
+  - Running multiple separate exo clusters on the same network
+  - Isolating development/testing clusters from production clusters
+  - Preventing accidental cluster joining
+
+- **Configuration**: Access this setting in the app's Advanced settings (or set the `EXO_LIBP2P_NAMESPACE` environment variable when running from source)
+
+The namespace is logged on startup for debugging purposes.
+
+#### Uninstalling the macOS App
+
+The recommended way to uninstall is through the app itself: click the menu bar icon → Advanced → Uninstall. This cleanly removes all system components.
+
+If you've already deleted the app, you can run the standalone uninstaller script:
+
+```bash
+sudo ./app/EXO/uninstall-exo.sh
+```
+
+This removes:
+- Network setup LaunchDaemon
+- Network configuration script
+- Log files
+- The "exo" network location
+
+**Note:** You'll need to manually remove EXO from Login Items in System Settings → General → Login Items.
+
 ---
 
 ### Enabling RDMA on macOS
@@ -287,7 +345,56 @@ curl -X DELETE http://localhost:52415/instance/YOUR_INSTANCE_ID
 - List all models: `curl http://localhost:52415/models`
 - Inspect instance IDs and deployment state: `curl http://localhost:52415/state`
 
-For further details, see API types and endpoints in [src/exo/master/api.py](src/exo/master/api.py).
+For further details, see:
+
+- API basic documentation in [docs/api.md](docs/api.md).
+- API types and endpoints in [src/exo/master/api.py](src/exo/master/api.py).
+
+---
+
+## Benchmarking
+
+The `exo-bench` tool measures model prefill and token generation speed across different placement configurations. This helps you optimize model performance and validate improvements.
+
+**Prerequisites:**
+- Nodes should be running with `uv run exo` before benchmarking
+- The tool uses the `/bench/chat/completions` endpoint
+
+**Basic usage:**
+
+```bash
+uv run bench/exo_bench.py \
+  --model llama-3.2-1b \
+  --pp 128,256,512 \
+  --tg 128,256
+```
+
+**Key parameters:**
+
+- `--model`: Model to benchmark (short ID or HuggingFace ID)
+- `--pp`: Prompt size hints (comma-separated integers)
+- `--tg`: Generation lengths (comma-separated integers)
+- `--max-nodes`: Limit placements to N nodes (default: 4)
+- `--instance-meta`: Filter by `ring`, `jaccl`, or `both` (default: both)
+- `--sharding`: Filter by `pipeline`, `tensor`, or `both` (default: both)
+- `--repeat`: Number of repetitions per configuration (default: 1)
+- `--warmup`: Warmup runs per placement (default: 0)
+- `--json-out`: Output file for results (default: bench/results.json)
+
+**Example with filters:**
+
+```bash
+uv run bench/exo_bench.py \
+  --model llama-3.2-1b \
+  --pp 128,512 \
+  --tg 128 \
+  --max-nodes 2 \
+  --sharding tensor \
+  --repeat 3 \
+  --json-out my-results.json
+```
+
+The tool outputs performance metrics including prompt tokens per second (prompt_tps), generation tokens per second (generation_tps), and peak memory usage for each configuration.
 
 ---
 
@@ -299,4 +406,4 @@ On macOS, exo uses the GPU. On Linux, exo currently runs on CPU. We are working
 
 ## Contributing
 
-See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines on how to contribute to exo.
+See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines on how to contribute to exo.

TODO.md

@@ -19,6 +19,7 @@
 25. Rethink retry logic
 26. Task cancellation. When API http request gets cancelled, it should cancel corresponding task.
 27. Log cleanup - per-module log filters and default to DEBUG log levels
+28. Validate RDMA connections with ibv_devinfo in the info gatherer
 
 Potential refactors:
 

app/EXO/EXO.xcodeproj/project.pbxproj

@@ -585,7 +585,7 @@
 			repositoryURL = "https://github.com/sparkle-project/Sparkle.git";
 			requirement = {
 				kind = upToNextMajorVersion;
-				minimumVersion = 2.8.1;
+				minimumVersion = 2.9.0-beta.1;
 			};
 		};
 /* End XCRemoteSwiftPackageReference section */

app/EXO/EXO.xcodeproj/project.xcworkspace/xcshareddata/swiftpm/Package.resolved

@@ -6,8 +6,8 @@
       "kind" : "remoteSourceControl",
       "location" : "https://github.com/sparkle-project/Sparkle.git",
       "state" : {
-        "revision" : "5581748cef2bae787496fe6d61139aebe0a451f6",
-        "version" : "2.8.1"
+        "revision" : "e641adb41915a8409895e2e30666aa64e487b637",
+        "version" : "2.9.0-beta.1"
       }
     }
   ],

app/EXO/EXO/ContentView.swift

@@ -18,10 +18,11 @@ struct ContentView: View {
     @State private var deletingInstanceIDs: Set<String> = []
     @State private var showAllNodes = false
     @State private var showAllInstances = false
+    @State private var showAdvanced = false
     @State private var showDebugInfo = false
     @State private var bugReportInFlight = false
     @State private var bugReportMessage: String?
-    @State private var showAdvancedOptions = false
+    @State private var uninstallInProgress = false
     @State private var pendingNamespace: String = ""
 
     var body: some View {
@@ -55,6 +56,11 @@ struct ContentView: View {
     }
 
     private var shouldShowLocalNetworkWarning: Bool {
+        // Show warning if local network is not working and EXO is running.
+        // The checker uses a longer timeout on first launch to allow time for
+        // the permission prompt, so this correctly handles both:
+        // 1. User denied permission on first launch
+        // 2. Permission broke after restart (macOS TCC bug)
         if case .notWorking = localNetworkChecker.status {
             return controller.status != .stopped
         }
@@ -255,13 +261,7 @@ struct ContentView: View {
                 Divider()
                     .padding(.vertical, 4)
             }
-            controlButton(title: "Check for Updates") {
-                updater.checkForUpdates()
-            }
-            .padding(.bottom, 8)
-            advancedOptionsSection
-                .padding(.bottom, 8)
-            debugSection
+            advancedSection
                 .padding(.bottom, 8)
             controlButton(title: "Quit", tint: .secondary) {
                 controller.stop()
@@ -270,6 +270,54 @@ struct ContentView: View {
         }
     }
 
+    private var advancedSection: some View {
+        VStack(alignment: .leading, spacing: 6) {
+            HStack {
+                Text("Advanced")
+                    .font(.caption)
+                    .foregroundColor(.secondary)
+                Spacer()
+                collapseButton(isExpanded: $showAdvanced)
+            }
+            .animation(nil, value: showAdvanced)
+            if showAdvanced {
+                VStack(alignment: .leading, spacing: 8) {
+                    VStack(alignment: .leading, spacing: 4) {
+                        Text("Cluster Namespace")
+                            .font(.caption2)
+                            .foregroundColor(.secondary)
+                        HStack {
+                            TextField("optional", text: $pendingNamespace)
+                                .textFieldStyle(.roundedBorder)
+                                .font(.caption2)
+                                .onAppear {
+                                    pendingNamespace = controller.customNamespace
+                                }
+                            Button("Save & Restart") {
+                                controller.customNamespace = pendingNamespace
+                                if controller.status == .running || controller.status == .starting {
+                                    controller.restart()
+                                }
+                            }
+                            .font(.caption2)
+                            .disabled(pendingNamespace == controller.customNamespace)
+                        }
+                    }
+                    HoverButton(title: "Check for Updates", small: true) {
+                        updater.checkForUpdates()
+                    }
+                    debugSection
+                    HoverButton(title: "Uninstall", tint: .red, small: true) {
+                        showUninstallConfirmationAlert()
+                    }
+                    .disabled(uninstallInProgress)
+                }
+                .transition(.opacity)
+            }
+        }
+        .animation(.easeInOut(duration: 0.25), value: showAdvanced)
+    }
+
     private func controlButton(title: String, tint: Color = .primary, action: @escaping () -> Void)
         -> some View
     {
@@ -396,57 +444,16 @@ struct ContentView: View {
         }
     }
 
-    private var advancedOptionsSection: some View {
-        VStack(alignment: .leading, spacing: 6) {
-            HStack {
-                Text("Advanced Options")
-                    .font(.caption)
-                    .foregroundColor(.secondary)
-                Spacer()
-                collapseButton(isExpanded: $showAdvancedOptions)
-            }
-            .animation(nil, value: showAdvancedOptions)
-            if showAdvancedOptions {
-                VStack(alignment: .leading, spacing: 8) {
-                    VStack(alignment: .leading, spacing: 4) {
-                        Text("Cluster Namespace")
-                            .font(.caption2)
-                            .foregroundColor(.secondary)
-                        HStack {
-                            TextField("optional", text: $pendingNamespace)
-                                .textFieldStyle(.roundedBorder)
-                                .font(.caption2)
-                                .onAppear {
-                                    pendingNamespace = controller.customNamespace
-                                }
-                            Button("Save & Restart") {
-                                controller.customNamespace = pendingNamespace
-                                if controller.status == .running || controller.status == .starting {
-                                    controller.restart()
-                                }
-                            }
-                            .font(.caption2)
-                            .disabled(pendingNamespace == controller.customNamespace)
-                        }
-
-                    }
-                }
-                .transition(.opacity)
-            }
-        }
-        .animation(.easeInOut(duration: 0.25), value: showAdvancedOptions)
-    }
-
     private var debugSection: some View {
-        VStack(alignment: .leading, spacing: 6) {
-            HStack {
-                Text("Debug Info")
-                    .font(.caption)
-                    .foregroundColor(.secondary)
-                Spacer()
-                collapseButton(isExpanded: $showDebugInfo)
+        VStack(alignment: .leading, spacing: 4) {
+            HoverButton(
+                title: "Debug Info",
+                tint: .primary,
+                trailingSystemImage: showDebugInfo ? "chevron.up" : "chevron.down",
+                small: true
+            ) {
+                showDebugInfo.toggle()
             }
-            .animation(nil, value: showDebugInfo)
             if showDebugInfo {
                 VStack(alignment: .leading, spacing: 4) {
                     Text("Version: \(buildTag)")
@@ -463,6 +470,7 @@ struct ContentView: View {
                     sendBugReportButton
                         .padding(.top, 6)
                 }
+                .padding(.leading, 8)
                 .transition(.opacity)
             }
         }
@@ -604,6 +612,88 @@ struct ContentView: View {
         bugReportInFlight = false
     }
 
+    private func showUninstallConfirmationAlert() {
+        let alert = NSAlert()
+        alert.messageText = "Uninstall EXO"
+        alert.informativeText = """
+            This will remove EXO and all its system components:
+
+            • Network configuration daemon
+            • Launch at login registration
+            • EXO network location
+
+            The app will be moved to Trash.
+            """
+        alert.alertStyle = .warning
+        alert.addButton(withTitle: "Uninstall")
+        alert.addButton(withTitle: "Cancel")
+
+        // Style the Uninstall button as destructive
+        if let uninstallButton = alert.buttons.first {
+            uninstallButton.hasDestructiveAction = true
+        }
+
+        let response = alert.runModal()
+        if response == .alertFirstButtonReturn {
+            performUninstall()
+        }
+    }
+
+    private func performUninstall() {
+        uninstallInProgress = true
+
+        // Stop EXO process first
+        controller.cancelPendingLaunch()
+        controller.stop()
+        stateService.stopPolling()
+
+        // Run the privileged uninstall on a background thread
+        // Using .utility QoS to avoid priority inversion with NSAppleScript's subprocess
+        DispatchQueue.global(qos: .utility).async {
+            do {
+                // Remove network setup daemon and components (requires admin privileges)
+                try NetworkSetupHelper.uninstall()
+
+                DispatchQueue.main.async {
+                    // Unregister from launch at login
+                    LaunchAtLoginHelper.disable()
+
+                    // Move app to trash
+                    self.moveAppToTrash()
+
+                    // Quit the app
+                    DispatchQueue.main.asyncAfter(deadline: .now() + 0.5) {
+                        NSApplication.shared.terminate(nil)
+                    }
+                }
+            } catch {
+                DispatchQueue.main.async {
+                    self.showErrorAlert(message: error.localizedDescription)
+                    self.uninstallInProgress = false
+                }
+            }
+        }
+    }
+
+    private func showErrorAlert(message: String) {
+        let alert = NSAlert()
+        alert.messageText = "Uninstall Failed"
+        alert.informativeText = message
+        alert.alertStyle = .critical
+        alert.addButton(withTitle: "OK")
+        alert.runModal()
+    }
+
+    private func moveAppToTrash() {
+        guard let appURL = Bundle.main.bundleURL as URL? else { return }
+        do {
+            try FileManager.default.trashItem(at: appURL, resultingItemURL: nil)
+        } catch {
+            // If we can't trash the app, that's OK - user can do it manually
+            // The important system components have already been cleaned up
+        }
+    }
+
     private var buildTag: String {
         Bundle.main.infoDictionary?["EXOBuildTag"] as? String ?? "unknown"
     }
@@ -617,14 +707,27 @@ private struct HoverButton: View {
     let title: String
     let tint: Color
     let trailingSystemImage: String?
+    let small: Bool
     let action: () -> Void
 
+    init(
+        title: String, tint: Color = .primary, trailingSystemImage: String? = nil,
+        small: Bool = false, action: @escaping () -> Void
+    ) {
+        self.title = title
+        self.tint = tint
+        self.trailingSystemImage = trailingSystemImage
+        self.small = small
+        self.action = action
+    }
+
     @State private var isHovering = false
 
     var body: some View {
         Button(action: action) {
             HStack {
                 Text(title)
+                    .font(small ? .caption : nil)
                 Spacer()
                 if let systemName = trailingSystemImage {
                     Image(systemName: systemName)
@@ -632,8 +735,8 @@ private struct HoverButton: View {
                 }
             }
             .frame(maxWidth: .infinity, alignment: .leading)
-            .padding(.vertical, 6)
-            .padding(.horizontal, 8)
+            .padding(.vertical, small ? 4 : 6)
+            .padding(.horizontal, small ? 6 : 8)
             .background(
                 RoundedRectangle(cornerRadius: 6)
                     .fill(

app/EXO/EXO/EXOApp.swift

@@ -132,6 +132,24 @@ struct EXOApp: App {
     }
 }
 
+/// Helper for managing EXO's launch-at-login registration
+enum LaunchAtLoginHelper {
+    private static let logger = Logger(subsystem: "io.exo.EXO", category: "LaunchAtLogin")
+
+    /// Unregisters EXO from launching at login
+    static func disable() {
+        guard SMAppService.mainApp.status == .enabled else { return }
+        do {
+            try SMAppService.mainApp.unregister()
+            logger.info("Unregistered EXO from launch at login")
+        } catch {
+            logger.error(
+                "Failed to unregister EXO from launch at login: \(error.localizedDescription, privacy: .public)"
+            )
+        }
+    }
+}
+
 final class SparkleUpdater: NSObject, ObservableObject {
     private let controller: SPUStandardUpdaterController
     private let delegateProxy: ExoUpdaterDelegate

app/EXO/EXO/Services/LocalNetworkChecker.swift

@@ -5,8 +5,8 @@ import os.log
 /// Checks if the app's local network permission is actually functional.
 ///
 /// macOS local network permission can appear enabled in System Preferences but not
-/// actually work after a restart. This service detects this by creating a UDP
-/// connection to the mDNS multicast address (224.0.0.251:5353).
+/// actually work after a restart. This service uses NWConnection to mDNS multicast
+/// to verify actual connectivity.
 @MainActor
 final class LocalNetworkChecker: ObservableObject {
     enum Status: Equatable {
@@ -35,30 +35,43 @@ final class LocalNetworkChecker: ObservableObject {
     }
 
     private static let logger = Logger(subsystem: "io.exo.EXO", category: "LocalNetworkChecker")
+    private static let hasCompletedInitialCheckKey = "LocalNetworkChecker.hasCompletedInitialCheck"
 
     @Published private(set) var status: Status = .unknown
-    @Published private(set) var lastConnectionState: String = "none"
 
     private var connection: NWConnection?
     private var checkTask: Task<Void, Never>?
 
+    /// Whether we've completed at least one check (stored in UserDefaults)
+    private var hasCompletedInitialCheck: Bool {
+        get { UserDefaults.standard.bool(forKey: Self.hasCompletedInitialCheckKey) }
+        set { UserDefaults.standard.set(newValue, forKey: Self.hasCompletedInitialCheckKey) }
+    }
+
     /// Checks if local network access is working.
     func check() {
         checkTask?.cancel()
         status = .checking
-        lastConnectionState = "connecting"
+
+        // Use longer timeout on first launch to allow time for permission prompt
+        let isFirstCheck = !hasCompletedInitialCheck
+        let timeout: UInt64 = isFirstCheck ? 30_000_000_000 : 3_000_000_000
 
         checkTask = Task { [weak self] in
             guard let self else { return }
-            let result = await self.performCheck()
+
+            Self.logger.info("Checking local network connectivity (first check: \(isFirstCheck))")
+            let result = await self.checkConnectivity(timeout: timeout)
             self.status = result
+            self.hasCompletedInitialCheck = true
+
             Self.logger.info("Local network check complete: \(result.displayText)")
         }
     }
 
-    private func performCheck() async -> Status {
-        Self.logger.info("Checking local network access via UDP multicast")
-
+    /// Checks connectivity using NWConnection to mDNS multicast.
+    /// The connection attempt triggers the permission prompt if not yet shown.
+    private func checkConnectivity(timeout: UInt64) async -> Status {
         connection?.cancel()
         connection = nil
 
@@ -84,22 +97,7 @@ final class LocalNetworkChecker: ObservableObject {
                 continuation.resume(returning: status)
             }
 
-            conn.stateUpdateHandler = { [weak self] state in
-                let stateStr: String
-                switch state {
-                case .setup: stateStr = "setup"
-                case .preparing: stateStr = "preparing"
-                case .ready: stateStr = "ready"
-                case .waiting(let e): stateStr = "waiting(\(e))"
-                case .failed(let e): stateStr = "failed(\(e))"
-                case .cancelled: stateStr = "cancelled"
-                @unknown default: stateStr = "unknown"
-                }
-
-                Task { @MainActor in
-                    self?.lastConnectionState = stateStr
-                }
-
+            conn.stateUpdateHandler = { state in
                 switch state {
                 case .ready:
                     resumeOnce(.working)
@@ -108,6 +106,7 @@ final class LocalNetworkChecker: ObservableObject {
                     if errorStr.contains("54") || errorStr.contains("ECONNRESET") {
                         resumeOnce(.notWorking(reason: "Connection blocked"))
                     }
+                // Otherwise keep waiting - might be showing permission prompt
                 case .failed(let error):
                     let errorStr = "\(error)"
                     if errorStr.contains("65") || errorStr.contains("EHOSTUNREACH")
@@ -127,7 +126,7 @@ final class LocalNetworkChecker: ObservableObject {
             conn.start(queue: .main)
 
             Task {
-                try? await Task.sleep(nanoseconds: 3_000_000_000)
+                try? await Task.sleep(nanoseconds: timeout)
                 let state = conn.state
                 switch state {
                 case .ready:

app/EXO/EXO/Services/NetworkSetupHelper.swift

@@ -6,7 +6,7 @@ enum NetworkSetupHelper {
     private static let logger = Logger(subsystem: "io.exo.EXO", category: "NetworkSetup")
     private static let daemonLabel = "io.exo.networksetup"
     private static let scriptDestination =
-        "/Library/Application Support/EXO/disable_bridge_enable_dhcp.sh"
+        "/Library/Application Support/EXO/disable_bridge.sh"
     private static let plistDestination = "/Library/LaunchDaemons/io.exo.networksetup.plist"
     private static let requiredStartInterval: Int = 1791
 
@@ -28,42 +28,14 @@ enum NetworkSetupHelper {
         # Remove Thunderbolt Bridge from VirtualNetworkInterfaces in preferences.plist
         /usr/libexec/PlistBuddy -c "Delete :VirtualNetworkInterfaces:Bridge:bridge0" "$PREFS" 2>/dev/null || true
 
-        networksetup -listlocations | grep -q exo || {
-          networksetup -createlocation exo
-        }
-
-        networksetup -switchtolocation exo
-        networksetup -listallhardwareports \\
-          | awk -F': ' '/Hardware Port: / {print $2}' \\
-          | while IFS=":" read -r name; do
-              case "$name" in
-                "Ethernet Adapter"*)
-                        ;;
-                "Thunderbolt Bridge")
-                        ;;
-                "Thunderbolt "*)
-                  networksetup -listallnetworkservices \\
-                    | grep -q "EXO $name" \\
-                      || networksetup -createnetworkservice "EXO $name" "$name" 2>/dev/null \\
-                      || continue
-                  networksetup -setdhcp "EXO $name"
-                        ;;
-                *)
-                  networksetup -listallnetworkservices \\
-                    | grep -q "$name" \\
-                      || networksetup -createnetworkservice "$name" "$name" 2>/dev/null \\
-                      || continue
-                        ;;
-              esac
-            done
-
         networksetup -listnetworkservices | grep -q "Thunderbolt Bridge" && {
           networksetup -setnetworkserviceenabled "Thunderbolt Bridge" off
         } || true
         """
 
     static func ensureLaunchDaemonInstalled() {
-        Task.detached {
+        // Use .utility priority to match NSAppleScript's internal QoS and avoid priority inversion
+        Task.detached(priority: .utility) {
             do {
                 if daemonAlreadyInstalled() {
                     return
@@ -78,11 +50,75 @@ enum NetworkSetupHelper {
         }
     }
 
+    /// Removes all EXO network setup components from the system.
+    /// This includes the LaunchDaemon, scripts, logs, and network location.
+    /// Requires admin privileges.
+    static func uninstall() throws {
+        let uninstallScript = makeUninstallScript()
+        try runShellAsAdmin(uninstallScript)
+        logger.info("EXO network setup components removed successfully")
+    }
+
+    /// Checks if there are any EXO network components installed that need cleanup
+    static func hasInstalledComponents() -> Bool {
+        let manager = FileManager.default
+        let scriptExists = manager.fileExists(atPath: scriptDestination)
+        let plistExists = manager.fileExists(atPath: plistDestination)
+        return scriptExists || plistExists
+    }
+
+    private static func makeUninstallScript() -> String {
+        """
+        set -euo pipefail
+
+        LABEL="\(daemonLabel)"
+        SCRIPT_DEST="\(scriptDestination)"
+        PLIST_DEST="\(plistDestination)"
+        LOG_OUT="/var/log/\(daemonLabel).log"
+        LOG_ERR="/var/log/\(daemonLabel).err.log"
+
+        # Unload the LaunchDaemon if running
+        launchctl bootout system/"$LABEL" 2>/dev/null || true
+
+        # Remove LaunchDaemon plist
+        rm -f "$PLIST_DEST"
+
+        # Remove the script and parent directory if empty
+        rm -f "$SCRIPT_DEST"
+        rmdir "$(dirname "$SCRIPT_DEST")" 2>/dev/null || true
+
+        # Remove log files
+        rm -f "$LOG_OUT" "$LOG_ERR"
+
+        # Switch back to Automatic network location
+        networksetup -switchtolocation Automatic 2>/dev/null || true
+
+        # Delete the exo network location if it exists
+        networksetup -listlocations | grep -q '^exo$' && {
+          networksetup -deletelocation exo 2>/dev/null || true
+        } || true
+
+        # Re-enable Thunderbolt Bridge if it exists
+        networksetup -listnetworkservices | grep -q "Thunderbolt Bridge" && {
+          networksetup -setnetworkserviceenabled "Thunderbolt Bridge" on 2>/dev/null || true
+        } || true
+
+        echo "EXO network components removed successfully"
+        """
+    }
+
     private static func daemonAlreadyInstalled() -> Bool {
         let manager = FileManager.default
         let scriptExists = manager.fileExists(atPath: scriptDestination)
         let plistExists = manager.fileExists(atPath: plistDestination)
         guard scriptExists, plistExists else { return false }
+        guard
+            let installedScript = try? String(contentsOfFile: scriptDestination, encoding: .utf8),
+            installedScript.trimmingCharacters(in: .whitespacesAndNewlines)
+                == setupScript.trimmingCharacters(in: .whitespacesAndNewlines)
+        else {
+            return false
+        }
         guard
             let data = try? Data(contentsOf: URL(fileURLWithPath: plistDestination)),
             let plist = try? PropertyListSerialization.propertyList(

app/EXO/EXO/ViewModels/InstanceViewModel.swift

@@ -57,7 +57,7 @@ struct InstanceViewModel: Identifiable, Equatable {
         case waiting
         case failed
         case idle
-        case unknown
+        case preparing
 
         var label: String {
             switch self {
@@ -68,7 +68,7 @@ struct InstanceViewModel: Identifiable, Equatable {
             case .waiting: return "Waiting"
             case .failed: return "Failed"
             case .idle: return "Idle"
-            case .unknown: return "Unknown"
+            case .preparing: return "Preparing"
             }
         }
     }
@@ -185,7 +185,7 @@ extension InstanceViewModel.State {
         } else if statuses.isEmpty {
             self = .idle
         } else {
-            self = .unknown
+            self = .preparing
         }
     }
 }

app/EXO/EXO/Views/InstanceRowView.swift

@@ -83,7 +83,7 @@ struct InstanceRowView: View {
         case .ready: return .teal
         case .waiting, .idle: return .gray
         case .failed: return .red
-        case .unknown: return .secondary
+        case .preparing: return .secondary
         }
     }
 

app/EXO/uninstall-exo.sh

@@ -0,0 +1,154 @@
+#!/usr/bin/env bash
+#
+# EXO Uninstaller Script
+#
+# This script removes all EXO system components that persist after deleting the app.
+# Run with: sudo ./uninstall-exo.sh
+#
+# Components removed:
+# - LaunchDaemon: /Library/LaunchDaemons/io.exo.networksetup.plist
+# - Network script: /Library/Application Support/EXO/
+# - Log files: /var/log/io.exo.networksetup.*
+# - Network location: "exo"
+# - Launch at login registration
+#
+
+set -euo pipefail
+
+LABEL="io.exo.networksetup"
+SCRIPT_DEST="/Library/Application Support/EXO/disable_bridge_enable_dhcp.sh"
+PLIST_DEST="/Library/LaunchDaemons/io.exo.networksetup.plist"
+LOG_OUT="/var/log/${LABEL}.log"
+LOG_ERR="/var/log/${LABEL}.err.log"
+APP_BUNDLE_ID="io.exo.EXO"
+
+# Colors for output
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+NC='\033[0m' # No Color
+
+echo_info() {
+    echo -e "${GREEN}[INFO]${NC} $1"
+}
+
+echo_warn() {
+    echo -e "${YELLOW}[WARN]${NC} $1"
+}
+
+echo_error() {
+    echo -e "${RED}[ERROR]${NC} $1"
+}
+
+# Check if running as root
+if [[ $EUID -ne 0 ]]; then
+    echo_error "This script must be run as root (use sudo)"
+    exit 1
+fi
+
+echo ""
+echo "========================================"
+echo "        EXO Uninstaller"
+echo "========================================"
+echo ""
+
+# Unload the LaunchDaemon if running
+echo_info "Stopping network setup daemon..."
+if launchctl list | grep -q "$LABEL"; then
+    launchctl bootout system/"$LABEL" 2>/dev/null || true
+    echo_info "Daemon stopped"
+else
+    echo_warn "Daemon was not running"
+fi
+
+# Remove LaunchDaemon plist
+if [[ -f "$PLIST_DEST" ]]; then
+    rm -f "$PLIST_DEST"
+    echo_info "Removed LaunchDaemon plist"
+else
+    echo_warn "LaunchDaemon plist not found (already removed?)"
+fi
+
+# Remove the script and parent directory
+if [[ -f "$SCRIPT_DEST" ]]; then
+    rm -f "$SCRIPT_DEST"
+    echo_info "Removed network setup script"
+else
+    echo_warn "Network setup script not found (already removed?)"
+fi
+
+# Remove EXO directory if empty
+if [[ -d "/Library/Application Support/EXO" ]]; then
+    rmdir "/Library/Application Support/EXO" 2>/dev/null && \
+        echo_info "Removed EXO support directory" || \
+        echo_warn "EXO support directory not empty, leaving in place"
+fi
+
+# Remove log files
+if [[ -f "$LOG_OUT" ]] || [[ -f "$LOG_ERR" ]]; then
+    rm -f "$LOG_OUT" "$LOG_ERR"
+    echo_info "Removed log files"
+else
+    echo_warn "Log files not found (already removed?)"
+fi
+
+# Switch back to Automatic network location
+echo_info "Restoring network configuration..."
+if networksetup -listlocations | grep -q "^Automatic$"; then
+    networksetup -switchtolocation Automatic 2>/dev/null || true
+    echo_info "Switched to Automatic network location"
+else
+    echo_warn "Automatic network location not found"
+fi
+
+# Delete the exo network location if it exists
+if networksetup -listlocations | grep -q "^exo$"; then
+    networksetup -deletelocation exo 2>/dev/null || true
+    echo_info "Deleted 'exo' network location"
+else
+    echo_warn "'exo' network location not found (already removed?)"
+fi
+
+# Re-enable Thunderbolt Bridge if it exists
+if networksetup -listnetworkservices 2>/dev/null | grep -q "Thunderbolt Bridge"; then
+    networksetup -setnetworkserviceenabled "Thunderbolt Bridge" on 2>/dev/null || true
+    echo_info "Re-enabled Thunderbolt Bridge"
+fi
+
+# Note about launch at login registration
+# SMAppService-based login items cannot be removed from a shell script.
+# They can only be unregistered from within the app itself or manually via System Settings.
+echo_warn "Launch at login must be removed manually:"
+echo_warn "  System Settings → General → Login Items → Remove EXO"
+
+# Check if EXO.app exists in common locations
+APP_FOUND=false
+for app_path in "/Applications/EXO.app" "$HOME/Applications/EXO.app"; do
+    if [[ -d "$app_path" ]]; then
+        if [[ "$APP_FOUND" == false ]]; then
+            echo ""
+            APP_FOUND=true
+        fi
+        echo_warn "EXO.app found at: $app_path"
+        echo_warn "You may want to move it to Trash manually."
+    fi
+done
+
+echo ""
+echo "========================================"
+echo_info "EXO uninstall complete!"
+echo "========================================"
+echo ""
+echo "The following have been removed:"
+echo "  • Network setup LaunchDaemon"
+echo "  • Network configuration script"
+echo "  • Log files"
+echo "  • 'exo' network location"
+echo ""
+echo "Your network has been restored to use the 'Automatic' location."
+echo "Thunderbolt Bridge has been re-enabled (if present)."
+echo ""
+echo "Manual step required:"
+echo "  Remove EXO from Login Items in System Settings → General → Login Items"
+echo ""
+

bench/exo_bench.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import argparse
+import contextlib
 import http.client
 import json
 import os
@@ -15,9 +16,6 @@ from urllib.parse import urlencode
 from loguru import logger
 from transformers import AutoTokenizer
 
-from exo.shared.models.model_cards import MODEL_CARDS
-from exo.shared.types.memory import Memory
-
 
 class ExoHttpError(RuntimeError):
     def __init__(self, status: int, reason: str, body_preview: str):
@@ -26,7 +24,7 @@ class ExoHttpError(RuntimeError):
 
 
 class ExoClient:
-    def __init__(self, host: str, port: int, timeout_s: float = 2400.0):
+    def __init__(self, host: str, port: int, timeout_s: float = 600.0):
         self.host = host
         self.port = port
         self.timeout_s = timeout_s
@@ -104,22 +102,46 @@ def runner_ready(runner: dict[str, Any]) -> bool:
     return "RunnerReady" in runner
 
 
+def runner_failed(runner: dict[str, Any]) -> bool:
+    return "RunnerFailed" in runner
+
+
+def get_runner_failed_message(runner: dict[str, Any]) -> str | None:
+    if "RunnerFailed" in runner:
+        return runner["RunnerFailed"].get("errorMessage")
+    return None
+
+
 def wait_for_instance_ready(
     client: ExoClient, instance_id: str, timeout: float = 24000.0
 ) -> None:
     start_time = time.time()
+    instance_existed = False
     while time.time() - start_time < timeout:
         state = client.request_json("GET", "/state")
         instances = state.get("instances", {})
 
         if instance_id not in instances:
+            if instance_existed:
+                # Instance was deleted after being created - likely due to runner failure
+                raise RuntimeError(
+                    f"Instance {instance_id} was deleted (runner may have failed)"
+                )
             time.sleep(0.1)
             continue
 
+        instance_existed = True
         instance = instances[instance_id]
         runner_ids = runner_ids_from_instance(instance)
         runners = state.get("runners", {})
 
+        # Check for failed runners first
+        for rid in runner_ids:
+            runner = runners.get(rid, {})
+            if runner_failed(runner):
+                error_msg = get_runner_failed_message(runner) or "Unknown error"
+                raise RuntimeError(f"Runner {rid} failed: {error_msg}")
+
         if all(runner_ready(runners.get(rid, {})) for rid in runner_ids):
             return
 
@@ -241,6 +263,9 @@ class PromptSizer:
             ids = tokenizer.apply_chat_template(
                 messages, tokenize=True, add_generation_prompt=True
             )
+            # Fix for transformers 5.x
+            if hasattr(ids, "input_ids"):
+                ids = ids.input_ids
             return int(len(ids))
 
         return count_fn
@@ -296,6 +321,12 @@ def main() -> int:
         default=4,
         help="Only consider placements using <= this many nodes.",
     )
+    ap.add_argument(
+        "--min-nodes",
+        type=int,
+        default=1,
+        help="Only consider placements using >= this many nodes.",
+    )
     ap.add_argument(
         "--instance-meta", choices=["ring", "jaccl", "both"], default="both"
     )
@@ -317,7 +348,7 @@ def main() -> int:
         help="Warmup runs per placement (uses first pp/tg).",
     )
     ap.add_argument(
-        "--timeout", type=float, default=2400.0, help="HTTP timeout (seconds)."
+        "--timeout", type=float, default=600.0, help="HTTP timeout (seconds)."
     )
     ap.add_argument(
         "--json-out",
@@ -396,7 +427,7 @@ def main() -> int:
         ):
             continue
 
-        if 0 < n <= args.max_nodes:
+        if args.min_nodes <= n <= args.max_nodes:
             selected.append(p)
 
     if not selected:
@@ -438,7 +469,13 @@ def main() -> int:
         )
 
         client.request_json("POST", "/instance", body={"instance": instance})
-        wait_for_instance_ready(client, instance_id)
+        try:
+            wait_for_instance_ready(client, instance_id)
+        except (RuntimeError, TimeoutError) as e:
+            logger.error(f"Failed to initialize placement: {e}")
+            with contextlib.suppress(ExoHttpError):
+                client.request_json("DELETE", f"/instance/{instance_id}")
+            continue
 
         time.sleep(1)
 
@@ -450,17 +487,17 @@ def main() -> int:
                 logger.debug(f"  warmup {i + 1}/{args.warmup} done")
 
             for pp in pp_list:
-                if (
-                    pp * n_nodes > 2048
-                    and "ring" in instance_meta.lower()
-                    and "tensor" in sharding.lower()
-                ):
-                    model_card = MODEL_CARDS[short_id]
-                    if model_card.metadata.storage_size > Memory.from_gb(10):
-                        logger.info(
-                            f"Skipping tensor ring as this is too slow for model of size {model_card.metadata.storage_size} on {n_nodes=}"
-                        )
-                        continue
+                # if (
+                #     pp * n_nodes > 2048
+                #     and "ring" in instance_meta.lower()
+                #     and "tensor" in sharding.lower()
+                # ):
+                #     model_card = MODEL_CARDS[short_id]
+                #     if model_card.metadata.storage_size > Memory.from_gb(10):
+                #         logger.info(
+                #             f"Skipping tensor ring as this is too slow for model of size {model_card.metadata.storage_size} on {n_nodes=}"
+                #         )
+                #         continue
                 for tg in tg_list:
                     runs: list[dict[str, Any]] = []
                     for r in range(args.repeat):

dashboard/dashboard.nix

@@ -0,0 +1,60 @@
+{ lib
+, config
+, dream2nix
+, ...
+}:
+let
+  # Read and parse the lock file
+  rawLockFile = builtins.fromJSON (builtins.readFile "${config.deps.dashboardSrc}/package-lock.json");
+
+  # For packages with bundleDependencies, filter out deps that are bundled
+  # (bundled deps are inside the tarball, not separate lockfile entries)
+  fixedPackages = lib.mapAttrs
+    (path: entry:
+      if entry ? bundleDependencies && entry.bundleDependencies != [ ]
+      then entry // {
+        dependencies = lib.filterAttrs
+          (name: _: !(lib.elem name entry.bundleDependencies))
+          (entry.dependencies or { });
+      }
+      else entry
+    )
+    (rawLockFile.packages or { });
+
+  fixedLockFile = rawLockFile // { packages = fixedPackages; };
+in
+{
+  imports = [
+    dream2nix.modules.dream2nix.nodejs-package-lock-v3
+    dream2nix.modules.dream2nix.nodejs-granular-v3
+  ];
+
+  name = "exo-dashboard";
+  version = "1.0.0";
+
+  mkDerivation = {
+    src = config.deps.dashboardSrc;
+
+    buildPhase = ''
+      runHook preBuild
+      npm run build
+      runHook postBuild
+    '';
+
+    installPhase = ''
+      runHook preInstall
+      cp -r build $out/build
+      runHook postInstall
+    '';
+  };
+
+  deps = { nixpkgs, ... }: {
+    inherit (nixpkgs) stdenv;
+    dashboardSrc = null; # Injected by parts.nix
+  };
+
+  nodejs-package-lock-v3 = {
+    # Don't use packageLockFile - provide the fixed lock content directly
+    packageLock = fixedLockFile;
+  };
+}

dashboard/package-lock.json

@@ -863,6 +863,7 @@
 			"integrity": "sha512-oH8tXw7EZnie8FdOWYrF7Yn4IKrqTFHhXvl8YxXxbKwTMcD/5NNCryUSEXRk2ZR4ojnub0P8rNrsVGHXWqIDtA==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"@standard-schema/spec": "^1.0.0",
 				"@sveltejs/acorn-typescript": "^1.0.5",
@@ -902,6 +903,7 @@
 			"integrity": "sha512-Y1Cs7hhTc+a5E9Va/xwKlAJoariQyHY+5zBgCZg4PFWNYQ1nMN9sjK1zhw1gK69DuqVP++sht/1GZg1aRwmAXQ==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"@sveltejs/vite-plugin-svelte-inspector": "^4.0.1",
 				"debug": "^4.4.1",
@@ -1518,6 +1520,7 @@
 			"integrity": "sha512-LCCV0HdSZZZb34qifBsyWlUmok6W7ouER+oQIGBScS8EsZsQbrtFTUrDX4hOl+CS6p7cnNC4td+qrSVGSCTUfQ==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"undici-types": "~6.21.0"
 			}
@@ -1527,6 +1530,7 @@
 			"resolved": "https://registry.npmjs.org/acorn/-/acorn-8.15.0.tgz",
 			"integrity": "sha512-NZyJarBfL7nWwIq+FDL6Zp/yHEhePMNnnJ0y3qfieCrmNvYct8uvtiV41UvlSe6apAfk0fY1FbWx+NwfmpvtTg==",
 			"license": "MIT",
+			"peer": true,
 			"bin": {
 				"acorn": "bin/acorn"
 			},
@@ -1939,6 +1943,7 @@
 			"integrity": "sha512-fmTRWbNMmsmWq6xJV8D19U/gw/bwrHfNXxrIN+HfZgnzqTHp9jOmKMhsTUjXOJnZOdZY9Q28y4yebKzqDKlxlQ==",
 			"dev": true,
 			"license": "ISC",
+			"peer": true,
 			"engines": {
 				"node": ">=12"
 			}
@@ -2646,6 +2651,7 @@
 			"integrity": "sha512-5gTmgEY/sqK6gFXLIsQNH19lWb4ebPDLA4SdLP7dsWkIXHWlG66oPuVvXSGFPppYZz8ZDZq0dYYrbHfBCVUb1Q==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"engines": {
 				"node": ">=12"
 			},
@@ -2833,6 +2839,7 @@
 			"resolved": "https://registry.npmjs.org/svelte/-/svelte-5.45.3.tgz",
 			"integrity": "sha512-ngKXNhNvwPzF43QqEhDOue7TQTrG09em1sd4HBxVF0Wr2gopAmdEWan+rgbdgK4fhBtSOTJO8bYU4chUG7VXZQ==",
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"@jridgewell/remapping": "^2.3.4",
 				"@jridgewell/sourcemap-codec": "^1.5.0",
@@ -2977,6 +2984,7 @@
 			"integrity": "sha512-jl1vZzPDinLr9eUt3J/t7V6FgNEw9QjvBPdysz9KfQDD41fQrC2Y4vKQdiaUpFT4bXlb1RHhLpp8wtm6M5TgSw==",
 			"dev": true,
 			"license": "Apache-2.0",
+			"peer": true,
 			"bin": {
 				"tsc": "bin/tsc",
 				"tsserver": "bin/tsserver"
@@ -2998,6 +3006,7 @@
 			"integrity": "sha512-+Oxm7q9hDoLMyJOYfUYBuHQo+dkAloi33apOPP56pzj+vsdJDzr+j1NISE5pyaAuKL4A3UD34qd0lx5+kfKp2g==",
 			"dev": true,
 			"license": "MIT",
+			"peer": true,
 			"dependencies": {
 				"esbuild": "^0.25.0",
 				"fdir": "^6.4.4",

dashboard/parts.nix

@@ -0,0 +1,44 @@
+{ inputs, ... }:
+{
+  perSystem =
+    { pkgs, lib, ... }:
+    let
+      # Filter source to only include dashboard directory
+      src = lib.cleanSourceWith {
+        src = inputs.self;
+        filter =
+          path: type:
+          let
+            baseName = builtins.baseNameOf path;
+            inDashboardDir =
+              (lib.hasInfix "/dashboard/" path)
+              || (lib.hasSuffix "/dashboard" (builtins.dirOf path))
+              || (baseName == "dashboard" && type == "directory");
+          in
+          inDashboardDir;
+      };
+
+      # Build the dashboard with dream2nix (includes node_modules in output)
+      dashboardFull = inputs.dream2nix.lib.evalModules {
+        packageSets.nixpkgs = pkgs;
+        modules = [
+          ./dashboard.nix
+          {
+            paths.projectRoot = inputs.self;
+            paths.projectRootFile = "flake.nix";
+            paths.package = inputs.self + "/dashboard";
+          }
+          # Inject the filtered source
+          {
+            deps.dashboardSrc = lib.mkForce "${src}/dashboard";
+          }
+        ];
+      };
+    in
+    {
+      # Extract just the static site from the full build
+      packages.dashboard = pkgs.runCommand "exo-dashboard" { } ''
+        cp -r ${dashboardFull}/build $out
+      '';
+    };
+}

dashboard/src/lib/components/ChatForm.svelte

@@ -60,12 +60,39 @@
 		return models;
 	});
 
-	// Auto-select the first available model if none is selected
+	// Track previous model IDs to detect newly added models (plain variable to avoid reactive loop)
+	let previousModelIds: Set<string> = new Set();
+
+	// Auto-select the first available model if none is selected, if current selection is stale, or if a new model is added
 	$effect(() => {
 		const models = availableModels();
-		if (models.length > 0 && !currentModel) {
-			setSelectedChatModel(models[0].id);
+		const currentModelIds = new Set(models.map(m => m.id));
+
+		if (models.length > 0) {
+			// Find newly added models (in current but not in previous)
+			const newModels = models.filter(m => !previousModelIds.has(m.id));
+
+			// If no model selected, select the first available
+			if (!currentModel) {
+				setSelectedChatModel(models[0].id);
+			}
+			// If current model is stale (no longer has a running instance), reset to first available
+			else if (!models.some(m => m.id === currentModel)) {
+				setSelectedChatModel(models[0].id);
+			}
+			// If a new model was just added, select it
+			else if (newModels.length > 0 && previousModelIds.size > 0) {
+				setSelectedChatModel(newModels[0].id);
+			}
+		} else {
+			// No instances running - clear the selected model
+			if (currentModel) {
+				setSelectedChatModel('');
+			}
 		}
+
+		// Update previous model IDs for next comparison
+		previousModelIds = currentModelIds;
 	});
 
 	function getInstanceModelId(instanceWrapped: unknown): string {

dashboard/src/lib/components/MarkdownContent.svelte

@@ -53,62 +53,285 @@
 	marked.use({ renderer });
 
 	/**
-	 * Preprocess LaTeX: convert \(...\) to $...$ and \[...\] to $$...$$
-	 * Also protect code blocks from LaTeX processing
+	 * Unescape HTML entities that marked may have escaped
+	 */
+	function unescapeHtmlEntities(text: string): string {
+		return text
+			.replace(/&lt;/g, '<')
+			.replace(/&gt;/g, '>')
+			.replace(/&amp;/g, '&')
+			.replace(/&quot;/g, '"')
+			.replace(/&#39;/g, "'");
+	}
+
+	// Storage for math expressions extracted before markdown processing
+	const mathExpressions: Map<string, { content: string; displayMode: boolean }> = new Map();
+	let mathCounter = 0;
+
+	// Storage for HTML snippets that need protection from markdown
+	const htmlSnippets: Map<string, string> = new Map();
+	let htmlCounter = 0;
+
+	// Use alphanumeric placeholders that won't be interpreted as HTML tags
+	const MATH_PLACEHOLDER_PREFIX = 'MATHPLACEHOLDER';
+	const CODE_PLACEHOLDER_PREFIX = 'CODEPLACEHOLDER';
+	const HTML_PLACEHOLDER_PREFIX = 'HTMLPLACEHOLDER';
+
+	/**
+	 * Preprocess LaTeX: extract math, handle LaTeX document commands, and protect content
 	 */
 	function preprocessLaTeX(text: string): string {
-		// Protect code blocks
+		// Reset storage
+		mathExpressions.clear();
+		mathCounter = 0;
+		htmlSnippets.clear();
+		htmlCounter = 0;
+
+		// Protect code blocks first
 		const codeBlocks: string[] = [];
 		let processed = text.replace(/```[\s\S]*?```|`[^`]+`/g, (match) => {
 			codeBlocks.push(match);
-			return `<<CODE_${codeBlocks.length - 1}>>`;
+			return `${CODE_PLACEHOLDER_PREFIX}${codeBlocks.length - 1}END`;
 		});
 
-		// Convert \(...\) to $...$
-		processed = processed.replace(/\\\((.+?)\\\)/g, '$$$1$');
-		
-		// Convert \[...\] to $$...$$
-		processed = processed.replace(/\\\[([\s\S]*?)\\\]/g, '$$$$$1$$$$');
+		// Remove LaTeX document commands
+		processed = processed.replace(/\\documentclass(\[[^\]]*\])?\{[^}]*\}/g, '');
+		processed = processed.replace(/\\usepackage(\[[^\]]*\])?\{[^}]*\}/g, '');
+		processed = processed.replace(/\\begin\{document\}/g, '');
+		processed = processed.replace(/\\end\{document\}/g, '');
+		processed = processed.replace(/\\maketitle/g, '');
+		processed = processed.replace(/\\title\{[^}]*\}/g, '');
+		processed = processed.replace(/\\author\{[^}]*\}/g, '');
+		processed = processed.replace(/\\date\{[^}]*\}/g, '');
+
+		// Remove \require{...} commands (MathJax-specific, not supported by KaTeX)
+		processed = processed.replace(/\$\\require\{[^}]*\}\$/g, '');
+		processed = processed.replace(/\\require\{[^}]*\}/g, '');
+
+		// Remove unsupported LaTeX commands/environments (tikzpicture, figure, center, etc.)
+		processed = processed.replace(/\\begin\{tikzpicture\}[\s\S]*?\\end\{tikzpicture\}/g, () => {
+			const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+			htmlSnippets.set(placeholder, '<div class="latex-diagram-placeholder"><span class="latex-diagram-icon">📐</span><span class="latex-diagram-text">Diagram</span></div>');
+			htmlCounter++;
+			return placeholder;
+		});
+		processed = processed.replace(/\\begin\{figure\}[\s\S]*?\\end\{figure\}/g, () => {
+			const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+			htmlSnippets.set(placeholder, '<div class="latex-diagram-placeholder"><span class="latex-diagram-icon">🖼️</span><span class="latex-diagram-text">Figure</span></div>');
+			htmlCounter++;
+			return placeholder;
+		});
+		// Strip center environment (layout only, no content change)
+		processed = processed.replace(/\\begin\{center\}/g, '');
+		processed = processed.replace(/\\end\{center\}/g, '');
+		// Strip other layout environments
+		processed = processed.replace(/\\begin\{flushleft\}/g, '');
+		processed = processed.replace(/\\end\{flushleft\}/g, '');
+		processed = processed.replace(/\\begin\{flushright\}/g, '');
+		processed = processed.replace(/\\end\{flushright\}/g, '');
+		processed = processed.replace(/\\label\{[^}]*\}/g, '');
+		processed = processed.replace(/\\caption\{[^}]*\}/g, '');
+
+		// Protect escaped dollar signs (e.g., \$50 should become $50, not LaTeX)
+		processed = processed.replace(/\\\$/g, 'ESCAPEDDOLLARPLACEHOLDER');
+
+		// Convert LaTeX math environments to display math (both bare and wrapped in $...$)
+		const mathEnvs = ['align', 'align\\*', 'equation', 'equation\\*', 'gather', 'gather\\*', 'multline', 'multline\\*', 'eqnarray', 'eqnarray\\*', 'array', 'matrix', 'pmatrix', 'bmatrix', 'vmatrix', 'cases'];
+		for (const env of mathEnvs) {
+			// Handle $\begin{env}...\end{env}$ (with dollar signs, possibly multiline)
+			const wrappedRegex = new RegExp(`\\$\\\\begin\\{${env}\\}(\\{[^}]*\\})?([\\s\\S]*?)\\\\end\\{${env}\\}\\$`, 'g');
+			processed = processed.replace(wrappedRegex, (_, args, content) => {
+				const cleanEnv = env.replace('\\*', '*');
+				const mathContent = `\\begin{${cleanEnv}}${args || ''}${content}\\end{${cleanEnv}}`;
+				const placeholder = `${MATH_PLACEHOLDER_PREFIX}DISPLAY${mathCounter}END`;
+				mathExpressions.set(placeholder, { content: mathContent, displayMode: true });
+				mathCounter++;
+				return placeholder;
+			});
+
+			// Handle bare \begin{env}...\end{env} (without dollar signs)
+			const bareRegex = new RegExp(`\\\\begin\\{${env}\\}(\\{[^}]*\\})?([\\s\\S]*?)\\\\end\\{${env}\\}`, 'g');
+			processed = processed.replace(bareRegex, (_, args, content) => {
+				const cleanEnv = env.replace('\\*', '*');
+				const mathContent = `\\begin{${cleanEnv}}${args || ''}${content}\\end{${cleanEnv}}`;
+				const placeholder = `${MATH_PLACEHOLDER_PREFIX}DISPLAY${mathCounter}END`;
+				mathExpressions.set(placeholder, { content: mathContent, displayMode: true });
+				mathCounter++;
+				return placeholder;
+			});
+		}
+
+		// Convert LaTeX proof environments to styled blocks (use placeholders for HTML)
+		processed = processed.replace(
+			/\\begin\{proof\}([\s\S]*?)\\end\{proof\}/g,
+			(_, content) => {
+				const html = `<div class="latex-proof"><div class="latex-proof-header">Proof</div><div class="latex-proof-content">${content}</div></div>`;
+				const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+				htmlSnippets.set(placeholder, html);
+				htmlCounter++;
+				return placeholder;
+			}
+		);
+
+		// Convert LaTeX theorem-like environments
+		const theoremEnvs = ['theorem', 'lemma', 'corollary', 'proposition', 'definition', 'remark', 'example'];
+		for (const env of theoremEnvs) {
+			const envRegex = new RegExp(`\\\\begin\\{${env}\\}([\\s\\S]*?)\\\\end\\{${env}\\}`, 'gi');
+			const envName = env.charAt(0).toUpperCase() + env.slice(1);
+			processed = processed.replace(envRegex, (_, content) => {
+				const html = `<div class="latex-theorem"><div class="latex-theorem-header">${envName}</div><div class="latex-theorem-content">${content}</div></div>`;
+				const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+				htmlSnippets.set(placeholder, html);
+				htmlCounter++;
+				return placeholder;
+			});
+		}
+
+		// Convert LaTeX text formatting commands (use placeholders to protect from markdown)
+		processed = processed.replace(/\\emph\{([^}]*)\}/g, (_, content) => {
+			const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+			htmlSnippets.set(placeholder, `<em>${content}</em>`);
+			htmlCounter++;
+			return placeholder;
+		});
+		processed = processed.replace(/\\textit\{([^}]*)\}/g, (_, content) => {
+			const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+			htmlSnippets.set(placeholder, `<em>${content}</em>`);
+			htmlCounter++;
+			return placeholder;
+		});
+		processed = processed.replace(/\\textbf\{([^}]*)\}/g, (_, content) => {
+			const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+			htmlSnippets.set(placeholder, `<strong>${content}</strong>`);
+			htmlCounter++;
+			return placeholder;
+		});
+		processed = processed.replace(/\\texttt\{([^}]*)\}/g, (_, content) => {
+			const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+			htmlSnippets.set(placeholder, `<code class="inline-code">${content}</code>`);
+			htmlCounter++;
+			return placeholder;
+		});
+		processed = processed.replace(/\\underline\{([^}]*)\}/g, (_, content) => {
+			const placeholder = `${HTML_PLACEHOLDER_PREFIX}${htmlCounter}END`;
+			htmlSnippets.set(placeholder, `<u>${content}</u>`);
+			htmlCounter++;
+			return placeholder;
+		});
+
+		// Handle LaTeX line breaks and spacing
+		processed = processed.replace(/\\\\(?:\s*\n)?/g, '\n'); // \\ -> newline
+		processed = processed.replace(/\\newline/g, '\n');
+		processed = processed.replace(/\\par\b/g, '\n\n');
+		processed = processed.replace(/\\quad/g, ' ');
+		processed = processed.replace(/\\qquad/g, '  ');
+		processed = processed.replace(/~~/g, ' '); // non-breaking space
+
+		// Remove other common LaTeX commands that don't render
+		processed = processed.replace(/\\centering/g, '');
+		processed = processed.replace(/\\noindent/g, '');
+		processed = processed.replace(/\\hfill/g, '');
+		processed = processed.replace(/\\vspace\{[^}]*\}/g, '');
+		processed = processed.replace(/\\hspace\{[^}]*\}/g, ' ');
+
+		// Convert \(...\) to placeholder (display: false)
+		processed = processed.replace(/\\\(([\s\S]+?)\\\)/g, (_, content) => {
+			const placeholder = `${MATH_PLACEHOLDER_PREFIX}INLINE${mathCounter}END`;
+			mathExpressions.set(placeholder, { content, displayMode: false });
+			mathCounter++;
+			return placeholder;
+		});
+
+		// Convert \[...\] to placeholder (display: true)
+		processed = processed.replace(/\\\[([\s\S]*?)\\\]/g, (_, content) => {
+			const placeholder = `${MATH_PLACEHOLDER_PREFIX}DISPLAY${mathCounter}END`;
+			mathExpressions.set(placeholder, { content, displayMode: true });
+			mathCounter++;
+			return placeholder;
+		});
+
+		// Extract display math ($$...$$) BEFORE markdown processing
+		processed = processed.replace(/\$\$([\s\S]*?)\$\$/g, (_, content) => {
+			const placeholder = `${MATH_PLACEHOLDER_PREFIX}DISPLAY${mathCounter}END`;
+			mathExpressions.set(placeholder, { content: content.trim(), displayMode: true });
+			mathCounter++;
+			return placeholder;
+		});
+
+		// Extract inline math ($...$) BEFORE markdown processing
+		// Allow single-line only, skip currency patterns like $5 or $50
+		processed = processed.replace(/\$([^\$\n]+?)\$/g, (match, content) => {
+			if (/^\d/.test(content.trim())) {
+				return match; // Keep as-is for currency
+			}
+			const placeholder = `${MATH_PLACEHOLDER_PREFIX}INLINE${mathCounter}END`;
+			mathExpressions.set(placeholder, { content: content.trim(), displayMode: false });
+			mathCounter++;
+			return placeholder;
+		});
+
+		// Restore escaped dollar signs
+		processed = processed.replace(/ESCAPEDDOLLARPLACEHOLDER/g, '$');
 
 		// Restore code blocks
-		processed = processed.replace(/<<CODE_(\d+)>>/g, (_, index) => codeBlocks[parseInt(index)]);
+		processed = processed.replace(new RegExp(`${CODE_PLACEHOLDER_PREFIX}(\\d+)END`, 'g'), (_, index) => codeBlocks[parseInt(index)]);
+
+		// Clean up any remaining stray backslashes from unrecognized commands
+		processed = processed.replace(/\\(?=[a-zA-Z])/g, ''); // Remove \ before letters (unrecognized commands)
 
 		return processed;
 	}
 
 	/**
-	 * Render math expressions with KaTeX after HTML is generated
+	 * Render math expressions with KaTeX and restore HTML placeholders
 	 */
 	function renderMath(html: string): string {
-		// Render display math ($$...$$)
-		html = html.replace(/\$\$([\s\S]*?)\$\$/g, (_, math) => {
-			try {
-				return katex.renderToString(math.trim(), {
-					displayMode: true,
-					throwOnError: false,
-					output: 'html'
-				});
-			} catch {
-				return `<span class="math-error">$$${math}$$</span>`;
-			}
-		});
+		// Replace all math placeholders with rendered KaTeX
+		for (const [placeholder, { content, displayMode }] of mathExpressions) {
+			const escapedPlaceholder = placeholder.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
+			const regex = new RegExp(escapedPlaceholder, 'g');
 
-		// Render inline math ($...$) but avoid matching currency like $5
-		html = html.replace(/\$([^\$\n]+?)\$/g, (match, math) => {
-			// Skip if it looks like currency ($ followed by number)
-			if (/^\d/.test(math.trim())) {
-				return match;
-			}
-			try {
-				return katex.renderToString(math.trim(), {
-					displayMode: false,
-					throwOnError: false,
-					output: 'html'
-				});
-			} catch {
-				return `<span class="math-error">$${math}$</span>`;
-			}
-		});
+			html = html.replace(regex, () => {
+				try {
+					const rendered = katex.renderToString(content, {
+						displayMode,
+						throwOnError: false,
+						output: 'html'
+					});
+
+					if (displayMode) {
+						return `
+							<div class="math-display-wrapper">
+								<div class="math-display-header">
+									<span class="math-label">LaTeX</span>
+									<button type="button" class="copy-math-btn" data-math-source="${encodeURIComponent(content)}" title="Copy LaTeX source">
+										<svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">
+											<rect width="14" height="14" x="8" y="8" rx="2" ry="2"/>
+											<path d="M4 16c-1.1 0-2-.9-2-2V4c0-1.1.9-2 2-2h10c1.1 0 2 .9 2 2"/>
+										</svg>
+									</button>
+								</div>
+								<div class="math-display-content">
+									${rendered}
+								</div>
+							</div>
+						`;
+					} else {
+						return `<span class="math-inline">${rendered}</span>`;
+					}
+				} catch {
+					const display = displayMode ? `$$${content}$$` : `$${content}$`;
+					return `<span class="math-error"><span class="math-error-icon">⚠</span> ${display}</span>`;
+				}
+			});
+		}
+
+		// Restore HTML placeholders (for \textbf, \emph, etc.)
+		for (const [placeholder, htmlContent] of htmlSnippets) {
+			const escapedPlaceholder = placeholder.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
+			const regex = new RegExp(escapedPlaceholder, 'g');
+			html = html.replace(regex, htmlContent);
+		}
 
 		return html;
 	}
@@ -154,16 +377,50 @@
 		}
 	}
 
+	async function handleMathCopyClick(event: Event) {
+		const target = event.currentTarget as HTMLButtonElement;
+		const encodedSource = target.getAttribute('data-math-source');
+		if (!encodedSource) return;
+
+		const source = decodeURIComponent(encodedSource);
+
+		try {
+			await navigator.clipboard.writeText(source);
+			// Show copied feedback
+			const originalHtml = target.innerHTML;
+			target.innerHTML = `
+				<svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">
+					<path d="M20 6L9 17l-5-5"/>
+				</svg>
+			`;
+			target.classList.add('copied');
+			setTimeout(() => {
+				target.innerHTML = originalHtml;
+				target.classList.remove('copied');
+			}, 2000);
+		} catch (error) {
+			console.error('Failed to copy math:', error);
+		}
+	}
+
 	function setupCopyButtons() {
 		if (!containerRef || !browser) return;
 
-		const buttons = containerRef.querySelectorAll<HTMLButtonElement>('.copy-code-btn');
-		for (const button of buttons) {
+		const codeButtons = containerRef.querySelectorAll<HTMLButtonElement>('.copy-code-btn');
+		for (const button of codeButtons) {
 			if (button.dataset.listenerBound !== 'true') {
 				button.dataset.listenerBound = 'true';
 				button.addEventListener('click', handleCopyClick);
 			}
 		}
+
+		const mathButtons = containerRef.querySelectorAll<HTMLButtonElement>('.copy-math-btn');
+		for (const button of mathButtons) {
+			if (button.dataset.listenerBound !== 'true') {
+				button.dataset.listenerBound = 'true';
+				button.addEventListener('click', handleMathCopyClick);
+			}
+		}
 	}
 
 	$effect(() => {
@@ -424,28 +681,290 @@
 		color: #60a5fa;
 	}
 
-	/* KaTeX math styling */
+	/* KaTeX math styling - Base */
 	.markdown-content :global(.katex) {
 		font-size: 1.1em;
+		color: oklch(0.9 0 0);
 	}
 
-	.markdown-content :global(.katex-display) {
+	/* Display math container wrapper */
+	.markdown-content :global(.math-display-wrapper) {
 		margin: 1rem 0;
+		border-radius: 0.5rem;
+		overflow: hidden;
+		border: 1px solid rgba(255, 215, 0, 0.15);
+		background: rgba(0, 0, 0, 0.3);
+		transition: border-color 0.2s ease, box-shadow 0.2s ease;
+	}
+
+	.markdown-content :global(.math-display-wrapper:hover) {
+		border-color: rgba(255, 215, 0, 0.25);
+		box-shadow: 0 0 12px rgba(255, 215, 0, 0.08);
+	}
+
+	/* Display math header - hidden by default, slides in on hover */
+	.markdown-content :global(.math-display-header) {
+		display: flex;
+		justify-content: space-between;
+		align-items: center;
+		padding: 0.375rem 0.75rem;
+		background: rgba(255, 215, 0, 0.03);
+		border-bottom: 1px solid rgba(255, 215, 0, 0.08);
+		opacity: 0;
+		max-height: 0;
+		padding-top: 0;
+		padding-bottom: 0;
+		overflow: hidden;
+		transition:
+			opacity 0.2s ease,
+			max-height 0.2s ease,
+			padding 0.2s ease;
+	}
+
+	.markdown-content :global(.math-display-wrapper:hover .math-display-header) {
+		opacity: 1;
+		max-height: 2.5rem;
+		padding: 0.375rem 0.75rem;
+	}
+
+	.markdown-content :global(.math-label) {
+		color: rgba(255, 215, 0, 0.7);
+		font-size: 0.65rem;
+		font-weight: 500;
+		text-transform: uppercase;
+		letter-spacing: 0.1em;
+		font-family: ui-monospace, SFMono-Regular, 'SF Mono', Monaco, Consolas, monospace;
+	}
+
+	.markdown-content :global(.copy-math-btn) {
+		display: flex;
+		align-items: center;
+		justify-content: center;
+		padding: 0.25rem;
+		background: transparent;
+		border: none;
+		color: var(--exo-light-gray, #9ca3af);
+		cursor: pointer;
+		transition: color 0.2s;
+		border-radius: 0.25rem;
+		opacity: 0;
+		transition:
+			color 0.2s,
+			opacity 0.15s ease;
+	}
+
+	.markdown-content :global(.math-display-wrapper:hover .copy-math-btn) {
+		opacity: 1;
+	}
+
+	.markdown-content :global(.copy-math-btn:hover) {
+		color: var(--exo-yellow, #ffd700);
+	}
+
+	.markdown-content :global(.copy-math-btn.copied) {
+		color: #22c55e;
+	}
+
+	/* Display math content area */
+	.markdown-content :global(.math-display-content) {
+		padding: 1rem 1.25rem;
 		overflow-x: auto;
 		overflow-y: hidden;
-		padding: 0.5rem 0;
 	}
 
-	.markdown-content :global(.katex-display > .katex) {
+	/* Custom scrollbar for math overflow */
+	.markdown-content :global(.math-display-content::-webkit-scrollbar) {
+		height: 6px;
+	}
+
+	.markdown-content :global(.math-display-content::-webkit-scrollbar-track) {
+		background: rgba(255, 255, 255, 0.05);
+		border-radius: 3px;
+	}
+
+	.markdown-content :global(.math-display-content::-webkit-scrollbar-thumb) {
+		background: rgba(255, 215, 0, 0.2);
+		border-radius: 3px;
+	}
+
+	.markdown-content :global(.math-display-content::-webkit-scrollbar-thumb:hover) {
+		background: rgba(255, 215, 0, 0.35);
+	}
+
+	.markdown-content :global(.math-display-content .katex-display) {
+		margin: 0;
+		padding: 0;
+	}
+
+	.markdown-content :global(.math-display-content .katex-display > .katex) {
 		text-align: center;
 	}
 
+	/* Inline math wrapper */
+	.markdown-content :global(.math-inline) {
+		display: inline;
+		padding: 0 0.125rem;
+		border-radius: 0.25rem;
+		transition: background-color 0.15s ease;
+	}
+
+	.markdown-content :global(.math-inline:hover) {
+		background: rgba(255, 215, 0, 0.05);
+	}
+
+	/* Dark theme KaTeX overrides */
+	.markdown-content :global(.katex .mord),
+	.markdown-content :global(.katex .minner),
+	.markdown-content :global(.katex .mop),
+	.markdown-content :global(.katex .mbin),
+	.markdown-content :global(.katex .mrel),
+	.markdown-content :global(.katex .mpunct) {
+		color: oklch(0.9 0 0);
+	}
+
+	/* Fraction lines and rules */
+	.markdown-content :global(.katex .frac-line),
+	.markdown-content :global(.katex .overline-line),
+	.markdown-content :global(.katex .underline-line),
+	.markdown-content :global(.katex .hline),
+	.markdown-content :global(.katex .rule) {
+		border-color: oklch(0.85 0 0) !important;
+		background: oklch(0.85 0 0);
+	}
+
+	/* Square roots and SVG elements */
+	.markdown-content :global(.katex .sqrt-line) {
+		border-color: oklch(0.85 0 0) !important;
+	}
+
+	.markdown-content :global(.katex svg) {
+		fill: oklch(0.85 0 0);
+		stroke: oklch(0.85 0 0);
+	}
+
+	.markdown-content :global(.katex svg path) {
+		stroke: oklch(0.85 0 0);
+	}
+
+	/* Delimiters (parentheses, brackets, braces) */
+	.markdown-content :global(.katex .delimsizing),
+	.markdown-content :global(.katex .delim-size1),
+	.markdown-content :global(.katex .delim-size2),
+	.markdown-content :global(.katex .delim-size3),
+	.markdown-content :global(.katex .delim-size4),
+	.markdown-content :global(.katex .mopen),
+	.markdown-content :global(.katex .mclose) {
+		color: oklch(0.75 0 0);
+	}
+
+	/* Math error styling */
 	.markdown-content :global(.math-error) {
+		display: inline-flex;
+		align-items: center;
+		gap: 0.375rem;
 		color: #f87171;
 		font-family: ui-monospace, SFMono-Regular, 'SF Mono', Monaco, Consolas, monospace;
 		font-size: 0.875em;
 		background: rgba(248, 113, 113, 0.1);
-		padding: 0.125rem 0.25rem;
+		padding: 0.25rem 0.5rem;
 		border-radius: 0.25rem;
+		border: 1px solid rgba(248, 113, 113, 0.2);
+	}
+
+	.markdown-content :global(.math-error-icon) {
+		font-size: 0.875em;
+		opacity: 0.9;
+	}
+
+	/* LaTeX proof environment */
+	.markdown-content :global(.latex-proof) {
+		margin: 1rem 0;
+		padding: 1rem 1.25rem;
+		background: rgba(255, 255, 255, 0.02);
+		border-left: 3px solid rgba(255, 215, 0, 0.4);
+		border-radius: 0 0.375rem 0.375rem 0;
+	}
+
+	.markdown-content :global(.latex-proof-header) {
+		font-weight: 600;
+		font-style: italic;
+		color: oklch(0.85 0 0);
+		margin-bottom: 0.5rem;
+	}
+
+	.markdown-content :global(.latex-proof-header::after) {
+		content: '.';
+	}
+
+	.markdown-content :global(.latex-proof-content) {
+		color: oklch(0.9 0 0);
+	}
+
+	.markdown-content :global(.latex-proof-content p:last-child) {
+		margin-bottom: 0;
+	}
+
+	/* QED symbol at end of proof */
+	.markdown-content :global(.latex-proof-content::after) {
+		content: '∎';
+		display: block;
+		text-align: right;
+		color: oklch(0.7 0 0);
+		margin-top: 0.5rem;
+	}
+
+	/* LaTeX theorem-like environments */
+	.markdown-content :global(.latex-theorem) {
+		margin: 1rem 0;
+		padding: 1rem 1.25rem;
+		background: rgba(255, 215, 0, 0.03);
+		border: 1px solid rgba(255, 215, 0, 0.15);
+		border-radius: 0.375rem;
+	}
+
+	.markdown-content :global(.latex-theorem-header) {
+		font-weight: 700;
+		color: var(--exo-yellow, #ffd700);
+		margin-bottom: 0.5rem;
+	}
+
+	.markdown-content :global(.latex-theorem-header::after) {
+		content: '.';
+	}
+
+	.markdown-content :global(.latex-theorem-content) {
+		color: oklch(0.9 0 0);
+		font-style: italic;
+	}
+
+	.markdown-content :global(.latex-theorem-content p:last-child) {
+		margin-bottom: 0;
+	}
+
+	/* LaTeX diagram/figure placeholder */
+	.markdown-content :global(.latex-diagram-placeholder) {
+		display: flex;
+		align-items: center;
+		justify-content: center;
+		gap: 0.5rem;
+		margin: 1rem 0;
+		padding: 1.5rem 2rem;
+		background: rgba(255, 255, 255, 0.02);
+		border: 1px dashed rgba(255, 215, 0, 0.25);
+		border-radius: 0.5rem;
+		color: rgba(255, 215, 0, 0.6);
+		font-size: 0.875rem;
+	}
+
+	.markdown-content :global(.latex-diagram-icon) {
+		font-size: 1.25rem;
+		opacity: 0.8;
+	}
+
+	.markdown-content :global(.latex-diagram-text) {
+		font-family: ui-monospace, SFMono-Regular, 'SF Mono', Monaco, Consolas, monospace;
+		font-size: 0.75rem;
+		text-transform: uppercase;
+		letter-spacing: 0.05em;
 	}
 </style>

dashboard/src/lib/components/ModelCard.svelte

@@ -197,7 +197,7 @@ function toggleNodeDetails(nodeId: string): void {
 	// Uses API preview data when available, falls back to local estimation
 	const placementPreview = $derived(() => {
 		const nodeArray = nodeList();
-		if (nodeArray.length === 0) return { nodes: [], canFit: false, totalAvailable: 0, error: null };
+		if (nodeArray.length === 0) return { nodes: [], canFit: false, totalAvailable: 0, topoWidth: 260, topoHeight: 90, error: null };
 		
 		const numNodes = nodeArray.length;
 		const iconSize = numNodes === 1 ? 50 : 36;

dashboard/src/lib/components/TopologyGraph.svelte

@@ -1,7 +1,7 @@
 <script lang="ts">
 	import { onMount, onDestroy } from 'svelte';
 	import * as d3 from 'd3';
-import { topologyData, isTopologyMinimized, debugMode } from '$lib/stores/app.svelte';
+import { topologyData, isTopologyMinimized, debugMode, type NodeInfo } from '$lib/stores/app.svelte';
 
 	interface Props {
 		class?: string;
@@ -24,14 +24,14 @@ function getNodeLabel(nodeId: string): string {
 
 function getInterfaceLabel(nodeId: string, ip?: string): { label: string; missing: boolean } {
 	if (!ip) return { label: '?', missing: true };
-	
+
 	// Strip port if present (e.g., "192.168.1.1:8080" -> "192.168.1.1")
 	const cleanIp = ip.includes(':') && !ip.includes('[') ? ip.split(':')[0] : ip;
-	
+
 	// Helper to check a node's interfaces
-	function checkNode(node: typeof data.nodes[string]): string | null {
+	function checkNode(node: NodeInfo | undefined): string | null {
 		if (!node) return null;
-		
+
 		const matchFromInterfaces = node.network_interfaces?.find((iface) =>
 			(iface.addresses || []).some((addr) => addr === cleanIp || addr === ip)
 		);
@@ -39,17 +39,19 @@ function getInterfaceLabel(nodeId: string, ip?: string): { label: string; missin
 			return matchFromInterfaces.name;
 		}
 
-		const mapped = node.ip_to_interface?.[cleanIp] || node.ip_to_interface?.[ip];
-		if (mapped && mapped.trim().length > 0) {
-			return mapped;
+		if (node.ip_to_interface) {
+			const mapped = node.ip_to_interface[cleanIp] || (ip ? node.ip_to_interface[ip] : undefined);
+			if (mapped && mapped.trim().length > 0) {
+				return mapped;
+			}
 		}
 		return null;
 	}
-	
+
 	// Try specified node first
 	const result = checkNode(data?.nodes?.[nodeId]);
 	if (result) return { label: result, missing: false };
-	
+
 	// Fallback: search all nodes for this IP
 	for (const [, otherNode] of Object.entries(data?.nodes || {})) {
 		const otherResult = checkNode(otherNode);
@@ -255,21 +257,24 @@ function wrapLine(text: string, maxLen: number): string[] {
 		const arrowsGroup = svg.append('g').attr('class', 'arrows-group');
 		const debugLabelsGroup = svg.append('g').attr('class', 'debug-edge-labels');
 
-		const pairMap = new Map<string, { a: string; b: string; aToB: boolean; bToA: boolean; connections: Array<{ from: string; to: string; ip: string; ifaceLabel: string; missingIface: boolean }> }>();
-		let debugEdgeLabels: Array<{ connections: typeof pairMap extends Map<string, infer V> ? V['connections'] : never; isLeft: boolean; isTop: boolean; mx: number; my: number }> | null = null;
+		type ConnectionInfo = { from: string; to: string; ip: string; ifaceLabel: string; missingIface: boolean };
+		type PairEntry = { a: string; b: string; aToB: boolean; bToA: boolean; connections: ConnectionInfo[] };
+		type DebugEdgeLabelEntry = { connections: ConnectionInfo[]; isLeft: boolean; isTop: boolean; mx: number; my: number };
+		const pairMap = new Map<string, PairEntry>();
+		const debugEdgeLabels: DebugEdgeLabelEntry[] = [];
 		edges.forEach(edge => {
 			if (!edge.source || !edge.target || edge.source === edge.target) return;
 			if (!positionById[edge.source] || !positionById[edge.target]) return;
-			
+
 			const a = edge.source < edge.target ? edge.source : edge.target;
 			const b = edge.source < edge.target ? edge.target : edge.source;
 			const key = `${a}|${b}`;
 			const entry = pairMap.get(key) || { a, b, aToB: false, bToA: false, connections: [] };
-			
+
 			if (edge.source === a) entry.aToB = true;
 			else entry.bToA = true;
 
-			const ip = edge.sendBackIp || edge.sendBackMultiaddr?.ip_address || '?';
+			const ip = edge.sendBackIp || '?';
 			const ifaceInfo = getInterfaceLabel(edge.source, ip);
 			entry.connections.push({
 				from: edge.source,
@@ -338,9 +343,8 @@ function wrapLine(text: string, maxLen: number): string[] {
 				// Determine which side of viewport based on edge midpoint
 				const isLeft = mx < centerX;
 				const isTop = my < safeCenterY;
-				
+
 				// Store for batch rendering after all edges processed
-				if (!debugEdgeLabels) debugEdgeLabels = [];
 				debugEdgeLabels.push({
 					connections: entry.connections,
 					isLeft,
@@ -381,32 +385,32 @@ function wrapLine(text: string, maxLen: number): string[] {
 			}
 			
 			// Group by quadrant: topLeft, topRight, bottomLeft, bottomRight
-			const quadrants: Record<string, typeof debugEdgeLabels> = {
+			const quadrants: Record<string, DebugEdgeLabelEntry[]> = {
 				topLeft: [],
 				topRight: [],
 				bottomLeft: [],
 				bottomRight: []
 			};
-			
+
 			debugEdgeLabels.forEach(edge => {
 				const key = (edge.isTop ? 'top' : 'bottom') + (edge.isLeft ? 'Left' : 'Right');
 				quadrants[key].push(edge);
 			});
-			
+
 			// Render each quadrant
-			Object.entries(quadrants).forEach(([quadrant, edges]) => {
-				if (edges.length === 0) return;
-				
+			Object.entries(quadrants).forEach(([quadrant, quadrantEdges]) => {
+				if (quadrantEdges.length === 0) return;
+
 				const isLeft = quadrant.includes('Left');
 				const isTop = quadrant.includes('top');
-				
+
 				let baseX = isLeft ? padding : width - padding;
 				let baseY = isTop ? padding : height - padding;
 				const textAnchor = isLeft ? 'start' : 'end';
-				
+
 				let currentY = baseY;
-				
-				edges.forEach(edge => {
+
+				quadrantEdges.forEach(edge => {
 					edge.connections.forEach(conn => {
 						const arrow = getArrow(conn.from, conn.to);
 						const label = `${arrow} ${conn.ip} ${conn.ifaceLabel}`;

dashboard/src/lib/stores/app.svelte.ts

@@ -99,20 +99,36 @@ interface RawNodeProfile {
 
 interface RawTopologyNode {
 	nodeId: string;
-	nodeProfile: RawNodeProfile;
+	nodeProfile?: RawNodeProfile;
 }
 
-interface RawTopologyConnection {
-	localNodeId: string;
-	sendBackNodeId: string;
-	sendBackMultiaddr?:
-		| { multiaddr?: string; address?: string; ip_address?: string }
-		| string;
+// New connection edge types from Python SocketConnection/RDMAConnection
+interface RawSocketConnection {
+	sinkMultiaddr?: {
+		address?: string;
+		// Multiaddr uses snake_case (no camelCase alias)
+		ip_address?: string;
+		ipAddress?: string; // fallback in case it changes
+		address_type?: string;
+		port?: number;
+	};
 }
 
+interface RawRDMAConnection {
+	sourceRdmaIface?: string;
+	sinkRdmaIface?: string;
+}
+
+type RawConnectionEdge = RawSocketConnection | RawRDMAConnection;
+
+// New nested mapping format: { source: { sink: [edge1, edge2, ...] } }
+type RawConnectionsMap = Record<string, Record<string, RawConnectionEdge[]>>;
+
 interface RawTopology {
-	nodes: RawTopologyNode[];
-	connections?: RawTopologyConnection[];
+	// nodes can be array of strings (node IDs) or array of objects with nodeId/nodeProfile
+	nodes: (string | RawTopologyNode)[];
+	// New nested mapping format
+	connections?: RawConnectionsMap;
 }
 
 type RawNodeProfiles = Record<string, RawNodeProfile>;
@@ -213,9 +229,18 @@ function transformTopology(
 	const nodes: Record<string, NodeInfo> = {};
 	const edges: TopologyEdge[] = [];
 
+	// Handle nodes - can be array of strings (node IDs) or array of objects with nodeId/nodeProfile
 	for (const node of raw.nodes || []) {
-		const mergedProfile = profiles?.[node.nodeId];
-		const profile = { ...(node.nodeProfile ?? {}), ...(mergedProfile ?? {}) };
+		// Determine the node ID - could be a string or an object with nodeId property
+		const nodeId = typeof node === "string" ? node : node.nodeId;
+		if (!nodeId) continue;
+
+		// Get the profile - from the separate profiles map or from the node object itself
+		const profileFromMap = profiles?.[nodeId];
+		const profileFromNode =
+			typeof node === "object" ? node.nodeProfile : undefined;
+		const profile = { ...(profileFromNode ?? {}), ...(profileFromMap ?? {}) };
+
 		const ramTotal = profile?.memory?.ramTotal?.inBytes ?? 0;
 		const ramAvailable = profile?.memory?.ramAvailable?.inBytes ?? 0;
 		const ramUsage = Math.max(ramTotal - ramAvailable, 0);
@@ -264,7 +289,7 @@ function transformTopology(
 			}
 		}
 
-		nodes[node.nodeId] = {
+		nodes[nodeId] = {
 			system_info: {
 				model_id: profile?.modelId ?? "Unknown",
 				chip: profile?.chipId,
@@ -292,29 +317,34 @@ function transformTopology(
 		};
 	}
 
-	for (const conn of raw.connections || []) {
-		if (!conn.localNodeId || !conn.sendBackNodeId) continue;
-		if (conn.localNodeId === conn.sendBackNodeId) continue;
-		if (!nodes[conn.localNodeId] || !nodes[conn.sendBackNodeId]) continue;
+	// Handle connections - nested mapping format { source: { sink: [edges] } }
+	const connections = raw.connections;
+	if (connections && typeof connections === "object") {
+		for (const [source, sinks] of Object.entries(connections)) {
+			if (!sinks || typeof sinks !== "object") continue;
+			for (const [sink, edgeList] of Object.entries(sinks)) {
+				if (!Array.isArray(edgeList)) continue;
+				for (const edge of edgeList) {
+					// Extract IP from SocketConnection (uses snake_case: ip_address)
+					let sendBackIp: string | undefined;
+					if (edge && typeof edge === "object" && "sinkMultiaddr" in edge) {
+						const multiaddr = edge.sinkMultiaddr;
+						if (multiaddr) {
+							// Try both snake_case (actual) and camelCase (in case it changes)
+							sendBackIp =
+								multiaddr.ip_address ||
+								multiaddr.ipAddress ||
+								extractIpFromMultiaddr(multiaddr.address);
+						}
+					}
+					// RDMAConnection (sourceRdmaIface/sinkRdmaIface) has no IP - edge just shows connection exists
 
-		let sendBackIp: string | undefined;
-		if (conn.sendBackMultiaddr) {
-			const multi = conn.sendBackMultiaddr;
-			if (typeof multi === "string") {
-				sendBackIp = extractIpFromMultiaddr(multi);
-			} else {
-				sendBackIp =
-					multi.ip_address ||
-					extractIpFromMultiaddr(multi.multiaddr) ||
-					extractIpFromMultiaddr(multi.address);
+					if (nodes[source] && nodes[sink] && source !== sink) {
+						edges.push({ source, target: sink, sendBackIp });
+					}
+				}
 			}
 		}
-
-		edges.push({
-			source: conn.localNodeId,
-			target: conn.sendBackNodeId,
-			sendBackIp,
-		});
 	}
 
 	return { nodes, edges };

dashboard/src/routes/+page.svelte

@@ -400,10 +400,8 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 				const errorText = await response.text();
 				console.error('Failed to launch instance:', errorText);
 			} else {
-				// Auto-select the launched model only if no model is currently selected
-				if (!selectedChatModel()) {
-					setSelectedChatModel(modelId);
-				}
+				// Always auto-select the newly launched model so the user chats to what they just launched
+				setSelectedChatModel(modelId);
 				
 				// Scroll to the bottom of instances container to show the new instance
 				// Use multiple attempts to ensure DOM has updated with the new instance
@@ -593,7 +591,7 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 		// Unwrap the instance
 		const [instanceTag, instance] = getTagged(instanceWrapped);
 		if (!instance || typeof instance !== 'object') {
-			return { isDownloading: false, progress: null, statusText: 'UNKNOWN', perNode: [] };
+			return { isDownloading: false, progress: null, statusText: 'PREPARING', perNode: [] };
 		}
 
 		const inst = instance as { shardAssignments?: { nodeToRunner?: Record<string, string>; runnerToShard?: Record<string, unknown>; modelId?: string } };
@@ -706,7 +704,7 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 	function deriveInstanceStatus(instanceWrapped: unknown): { statusText: string; statusClass: string } {
 		const [, instance] = getTagged(instanceWrapped);
 		if (!instance || typeof instance !== 'object') {
-			return { statusText: 'UNKNOWN', statusClass: 'inactive' };
+			return { statusText: 'PREPARING', statusClass: 'inactive' };
 		}
 		
 		const inst = instance as { shardAssignments?: { runnerToShard?: Record<string, unknown> } };
@@ -735,7 +733,7 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 
 		const has = (s: string) => statuses.includes(s);
 
-		if (statuses.length === 0) return { statusText: 'UNKNOWN', statusClass: 'inactive' };
+		if (statuses.length === 0) return { statusText: 'PREPARING', statusClass: 'inactive' };
 		if (has('Failed')) return { statusText: 'FAILED', statusClass: 'failed' };
 		if (has('Shutdown')) return { statusText: 'SHUTDOWN', statusClass: 'inactive' };
 		if (has('Loading')) return { statusText: 'LOADING', statusClass: 'starting' };
@@ -763,6 +761,10 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 	async function deleteInstance(instanceId: string) {
 		if (!confirm(`Delete instance ${instanceId.slice(0, 8)}...?`)) return;
 		
+		// Get the model ID of the instance being deleted before we delete it
+		const deletedInstanceModelId = getInstanceModelId(instanceData[instanceId]);
+		const wasSelected = selectedChatModel() === deletedInstanceModelId;
+		
 		try {
 			const response = await fetch(`/instance/${instanceId}`, {
 				method: 'DELETE',
@@ -771,6 +773,24 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 			
 			if (!response.ok) {
 				console.error('Failed to delete instance:', response.status);
+			} else if (wasSelected) {
+				// If we deleted the currently selected model, switch to another available model
+				// Find another instance that isn't the one we just deleted
+				const remainingInstances = Object.entries(instanceData).filter(([id]) => id !== instanceId);
+				if (remainingInstances.length > 0) {
+					// Select the last instance (most recently added, since objects preserve insertion order)
+					const [, lastInstance] = remainingInstances[remainingInstances.length - 1];
+					const newModelId = getInstanceModelId(lastInstance);
+					if (newModelId && newModelId !== 'Unknown' && newModelId !== 'Unknown Model') {
+						setSelectedChatModel(newModelId);
+					} else {
+						// Clear selection if no valid model found
+						setSelectedChatModel('');
+					}
+				} else {
+					// No more instances, clear the selection
+					setSelectedChatModel('');
+				}
 			}
 		} catch (error) {
 			console.error('Error deleting instance:', error);
@@ -895,7 +915,7 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 		const runnerEntries = Object.entries(runnerToShard).map(([runnerId, shardWrapped]) => {
 			const [tag, shard] = getTagged(shardWrapped);
 			const meta = (shard as { modelMeta?: { worldSize?: number; nLayers?: number; deviceRank?: number } } | undefined);
-			const deviceRank = (meta?.deviceRank as number | undefined) ?? 0;
+			const deviceRank = meta?.modelMeta?.deviceRank ?? 0;
 			return { runnerId, tag, deviceRank };
 		});
 
@@ -1267,9 +1287,9 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 							<div class="flex-1 h-px bg-gradient-to-r from-exo-yellow/30 to-transparent"></div>
 						</div>
 						
-						<div 
+						<div
 							bind:this={instancesContainerRef}
-							class="max-h-72 space-y-3 overflow-y-auto"
+							class="max-h-72 xl:max-h-96 space-y-3 overflow-y-auto overflow-x-hidden py-px"
 						>
 								{#each Object.entries(instanceData) as [id, instance]}
 									{@const downloadInfo = getInstanceDownloadStatus(id, instance)}
@@ -1773,7 +1793,7 @@ function toggleInstanceDownloadDetails(nodeId: string): void {
 								<h3 class="text-xs text-exo-yellow font-mono tracking-[0.2em] uppercase">Instances</h3>
 								<div class="flex-1 h-px bg-gradient-to-r from-exo-yellow/30 to-transparent"></div>
 							</div>
-								<div class="space-y-3 max-h-72 overflow-y-auto pr-1">
+								<div class="space-y-3 max-h-72 xl:max-h-96 overflow-y-auto overflow-x-hidden py-px pr-1">
 									{#each Object.entries(instanceData) as [id, instance]}
 										{@const downloadInfo = getInstanceDownloadStatus(id, instance)}
 										{@const statusText = downloadInfo.statusText}

dashboard/src/routes/downloads/+page.svelte

@@ -199,7 +199,13 @@
 					const rawProgress = (downloadPayload as Record<string, unknown>).download_progress
 						?? (downloadPayload as Record<string, unknown>).downloadProgress
 						?? {};
-					const totalBytes = getBytes((rawProgress as Record<string, unknown>).total_bytes ?? (rawProgress as Record<string, unknown>).totalBytes);
+					// For DownloadCompleted, total_bytes is at top level; for DownloadOngoing, it's inside download_progress
+					const totalBytes = getBytes(
+						(downloadPayload as Record<string, unknown>).total_bytes
+						?? (downloadPayload as Record<string, unknown>).totalBytes
+						?? (rawProgress as Record<string, unknown>).total_bytes
+						?? (rawProgress as Record<string, unknown>).totalBytes
+					);
 					const downloadedBytes = getBytes((rawProgress as Record<string, unknown>).downloaded_bytes ?? (rawProgress as Record<string, unknown>).downloadedBytes);
 					const speed = (rawProgress as Record<string, unknown>).speed as number ?? 0;
 					const etaMs = (rawProgress as Record<string, unknown>).eta_ms as number ?? (rawProgress as Record<string, unknown>).etaMs as number ?? 0;
@@ -332,8 +338,13 @@
 								<div class="text-lg font-mono text-white truncate">{node.nodeName}</div>
 								<div class="text-xs text-exo-light-gray font-mono truncate">{node.nodeId}</div>
 							</div>
-							<div class="text-xs font-mono uppercase tracking-wider whitespace-nowrap shrink-0">
-								<span class="text-green-400">{node.models.filter(m => m.status === 'completed').length}</span><span class="text-exo-yellow"> /{node.models.length} models</span>
+							<div class="text-xs font-mono uppercase tracking-wider whitespace-nowrap shrink-0 text-right">
+								<div>
+									<span class="text-green-400">{node.models.filter(m => m.status === 'completed').length}</span><span class="text-exo-yellow"> / {node.models.length} models</span>
+								</div>
+								<div class="text-exo-light-gray normal-case tracking-normal">
+									{formatBytes(node.models.filter(m => m.status === 'completed').reduce((sum, m) => sum + m.totalBytes, 0))} on disk
+								</div>
 							</div>
 						</div>
 
@@ -385,7 +396,7 @@
 								</div>
 
 								<div class="flex items-center justify-between text-xs font-mono text-exo-light-gray">
-									<span>{model.status === 'completed' ? 'Completed' : `${formatSpeed(model.speed)} • ETA ${formatEta(model.etaMs)}`}</span>
+									<span>{model.status === 'completed' ? `Completed (${formatBytes(model.totalBytes)})` : `${formatSpeed(model.speed)} • ETA ${formatEta(model.etaMs)}`}</span>
 									{#if model.status !== 'completed'}
 										<span>{model.files.length} file{model.files.length === 1 ? '' : 's'}</span>
 									{/if}

docs/api.md

@@ -0,0 +1,212 @@
+# EXO API – Technical Reference
+
+This document describes the REST API exposed by the **EXO ** service, as implemented in:
+
+`src/exo/master/api.py`
+
+The API is used to manage model instances in the cluster, inspect cluster state, and perform inference using an OpenAI-compatible interface.
+
+Base URL example:
+
+```
+http://localhost:52415
+```
+
+## 1. General / Meta Endpoints
+
+### Get Master Node ID
+
+**GET** `/node_id`
+
+Returns the identifier of the current master node.
+
+**Response (example):**
+
+```json
+{
+  "node_id": "node-1234"
+}
+```
+
+### Get Cluster State
+
+**GET** `/state`
+
+Returns the current state of the cluster, including nodes and active instances.
+
+**Response:**
+JSON object describing topology, nodes, and instances.
+
+### Get Events
+
+**GET** `/events`
+
+Returns the list of internal events recorded by the master (mainly for debugging and observability).
+
+**Response:**
+Array of event objects.
+
+## 2. Model Instance Management
+
+### Create Instance
+
+**POST** `/instance`
+
+Creates a new model instance in the cluster.
+
+**Request body (example):**
+
+```json
+{
+  "instance": {
+    "model_id": "llama-3.2-1b",
+    "placement": { }
+  }
+}
+```
+
+**Response:**
+JSON description of the created instance.
+
+### Delete Instance
+
+**DELETE** `/instance/{instance_id}`
+
+Deletes an existing instance by ID.
+
+**Path parameters:**
+
+* `instance_id`: string, ID of the instance to delete
+
+**Response:**
+Status / confirmation JSON.
+
+### Get Instance
+
+**GET** `/instance/{instance_id}`
+
+Returns details of a specific instance.
+
+**Path parameters:**
+
+* `instance_id`: string
+
+**Response:**
+JSON description of the instance.
+
+### Preview Placements
+
+**GET** `/instance/previews?model_id=...`
+
+Returns possible placement previews for a given model.
+
+**Query parameters:**
+
+* `model_id`: string, required
+
+**Response:**
+Array of placement preview objects.
+
+### Compute Placement
+
+**GET** `/instance/placement`
+
+Computes a placement for a potential instance without creating it.
+
+**Query parameters (typical):**
+
+* `model_id`: string
+* `sharding`: string or config
+* `instance_meta`: JSON-encoded metadata
+* `min_nodes`: integer
+
+**Response:**
+JSON object describing the proposed placement / instance configuration.
+
+### Place Instance (Dry Operation)
+
+**POST** `/place_instance`
+
+Performs a placement operation for an instance (planning step), without necessarily creating it.
+
+**Request body:**
+JSON describing the instance to be placed.
+
+**Response:**
+Placement result.
+
+## 3. Models
+
+### List Models
+
+**GET** `/models`
+**GET** `/v1/models` (alias)
+
+Returns the list of available models and their metadata.
+
+**Response:**
+Array of model descriptors.
+
+## 4. Inference / Chat Completions
+
+### OpenAI-Compatible Chat Completions
+
+**POST** `/v1/chat/completions`
+
+Executes a chat completion request using an OpenAI-compatible schema. Supports streaming and non-streaming modes.
+
+**Request body (example):**
+
+```json
+{
+  "model": "llama-3.2-1b",
+  "messages": [
+    { "role": "system", "content": "You are a helpful assistant." },
+    { "role": "user", "content": "Hello" }
+  ],
+  "stream": false
+}
+```
+
+**Response:**
+OpenAI-compatible chat completion response.
+
+### Benchmarked Chat Completions
+
+**POST** `/bench/chat/completions`
+
+Same as `/v1/chat/completions`, but also returns performance and generation statistics.
+
+**Request body:**
+Same schema as `/v1/chat/completions`.
+
+**Response:**
+Chat completion plus benchmarking metrics.
+
+## 5. Complete Endpoint Summary
+
+```
+GET     /node_id
+GET     /state
+GET     /events
+
+POST    /instance
+GET     /instance/{instance_id}
+DELETE  /instance/{instance_id}
+
+GET     /instance/previews
+GET     /instance/placement
+POST    /place_instance
+
+GET     /models
+GET     /v1/models
+
+POST    /v1/chat/completions
+POST    /bench/chat/completions
+```
+
+## 6. Notes
+
+* The `/v1/chat/completions` endpoint is compatible with the OpenAI API format, so existing OpenAI clients can be pointed to EXO by changing the base URL.
+* The instance placement endpoints allow you to plan and preview cluster allocations before actually creating instances.
+* The `/events` and `/state` endpoints are primarily intended for operational visibility and debugging.

flake.lock

@@ -1,5 +1,42 @@
 {
   "nodes": {
+    "crane": {
+      "locked": {
+        "lastModified": 1767744144,
+        "narHash": "sha256-9/9ntI0D+HbN4G0TrK3KmHbTvwgswz7p8IEJsWyef8Q=",
+        "owner": "ipetkov",
+        "repo": "crane",
+        "rev": "2fb033290bf6b23f226d4c8b32f7f7a16b043d7e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "ipetkov",
+        "repo": "crane",
+        "type": "github"
+      }
+    },
+    "dream2nix": {
+      "inputs": {
+        "nixpkgs": [
+          "nixpkgs"
+        ],
+        "purescript-overlay": "purescript-overlay",
+        "pyproject-nix": "pyproject-nix"
+      },
+      "locked": {
+        "lastModified": 1765953015,
+        "narHash": "sha256-5FBZbbWR1Csp3Y2icfRkxMJw/a/5FGg8hCXej2//bbI=",
+        "owner": "nix-community",
+        "repo": "dream2nix",
+        "rev": "69eb01fa0995e1e90add49d8ca5bcba213b0416f",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-community",
+        "repo": "dream2nix",
+        "type": "github"
+      }
+    },
     "fenix": {
       "inputs": {
         "nixpkgs": [
@@ -8,11 +45,11 @@
         "rust-analyzer-src": "rust-analyzer-src"
       },
       "locked": {
-        "lastModified": 1761893049,
-        "narHash": "sha256-1TtFDPhC+ZsrOOtBnry1EZC+WipTTvsOVjIEVugqji8=",
+        "lastModified": 1768287139,
+        "narHash": "sha256-nsXFt0OzUi6K7dUzzJD5/v9e0Ic+fvclfIW936/43ZM=",
         "owner": "nix-community",
         "repo": "fenix",
-        "rev": "c2ac9a5c0d6d16630c3b225b874bd14528d1abe6",
+        "rev": "a4a3aa956931f90f35453cb519e4545e9ad7f773",
         "type": "github"
       },
       "original": {
@@ -21,25 +58,59 @@
         "type": "github"
       }
     },
-    "flake-utils": {
-      "inputs": {
-        "systems": "systems"
-      },
+    "flake-compat": {
+      "flake": false,
       "locked": {
-        "lastModified": 1731533236,
-        "narHash": "sha256-l0KFg5HjrsfsO/JpG+r7fRrqm12kzFHyUHqHCVpMMbI=",
-        "owner": "numtide",
-        "repo": "flake-utils",
-        "rev": "11707dc2f618dd54ca8739b309ec4fc024de578b",
+        "lastModified": 1696426674,
+        "narHash": "sha256-kvjfFW7WAETZlt09AgDn1MrtKzP7t90Vf7vypd3OL1U=",
+        "owner": "edolstra",
+        "repo": "flake-compat",
+        "rev": "0f9255e01c2351cc7d116c072cb317785dd33b33",
         "type": "github"
       },
       "original": {
-        "owner": "numtide",
-        "repo": "flake-utils",
+        "owner": "edolstra",
+        "repo": "flake-compat",
+        "type": "github"
+      }
+    },
+    "flake-parts": {
+      "inputs": {
+        "nixpkgs-lib": [
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1768135262,
+        "narHash": "sha256-PVvu7OqHBGWN16zSi6tEmPwwHQ4rLPU9Plvs8/1TUBY=",
+        "owner": "hercules-ci",
+        "repo": "flake-parts",
+        "rev": "80daad04eddbbf5a4d883996a73f3f542fa437ac",
+        "type": "github"
+      },
+      "original": {
+        "owner": "hercules-ci",
+        "repo": "flake-parts",
         "type": "github"
       }
     },
     "nixpkgs": {
+      "locked": {
+        "lastModified": 1768127708,
+        "narHash": "sha256-1Sm77VfZh3mU0F5OqKABNLWxOuDeHIlcFjsXeeiPazs=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "ffbc9f8cbaacfb331b6017d5a5abb21a492c9a38",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "ref": "nixos-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "nixpkgs-swift": {
       "locked": {
         "lastModified": 1761672384,
         "narHash": "sha256-o9KF3DJL7g7iYMZq9SWgfS1BFlNbsm6xplRjVlOCkXI=",
@@ -50,27 +121,74 @@
       },
       "original": {
         "owner": "NixOS",
-        "ref": "nixos-unstable",
         "repo": "nixpkgs",
+        "rev": "08dacfca559e1d7da38f3cf05f1f45ee9bfd213c",
+        "type": "github"
+      }
+    },
+    "purescript-overlay": {
+      "inputs": {
+        "flake-compat": "flake-compat",
+        "nixpkgs": [
+          "dream2nix",
+          "nixpkgs"
+        ],
+        "slimlock": "slimlock"
+      },
+      "locked": {
+        "lastModified": 1728546539,
+        "narHash": "sha256-Sws7w0tlnjD+Bjck1nv29NjC5DbL6nH5auL9Ex9Iz2A=",
+        "owner": "thomashoneyman",
+        "repo": "purescript-overlay",
+        "rev": "4ad4c15d07bd899d7346b331f377606631eb0ee4",
+        "type": "github"
+      },
+      "original": {
+        "owner": "thomashoneyman",
+        "repo": "purescript-overlay",
+        "type": "github"
+      }
+    },
+    "pyproject-nix": {
+      "inputs": {
+        "nixpkgs": [
+          "dream2nix",
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1763017646,
+        "narHash": "sha256-Z+R2lveIp6Skn1VPH3taQIuMhABg1IizJd8oVdmdHsQ=",
+        "owner": "pyproject-nix",
+        "repo": "pyproject.nix",
+        "rev": "47bd6f296502842643078d66128f7b5e5370790c",
+        "type": "github"
+      },
+      "original": {
+        "owner": "pyproject-nix",
+        "repo": "pyproject.nix",
         "type": "github"
       }
     },
     "root": {
       "inputs": {
+        "crane": "crane",
+        "dream2nix": "dream2nix",
         "fenix": "fenix",
-        "flake-utils": "flake-utils",
+        "flake-parts": "flake-parts",
         "nixpkgs": "nixpkgs",
+        "nixpkgs-swift": "nixpkgs-swift",
         "treefmt-nix": "treefmt-nix"
       }
     },
     "rust-analyzer-src": {
       "flake": false,
       "locked": {
-        "lastModified": 1761849405,
-        "narHash": "sha256-igXdvC+WCUN+3gnfk+ptT7rMmxQuY6WbIg1rXMUN1DM=",
+        "lastModified": 1768224240,
+        "narHash": "sha256-Pp1dDrXKPBUJReZnnDElFyHYn67XTd48zRhToheLjtk=",
         "owner": "rust-lang",
         "repo": "rust-analyzer",
-        "rev": "f7de8ae045a5fe80f1203c5a1c3015b05f7c3550",
+        "rev": "725349602e525df37f377701e001fe8aab807878",
         "type": "github"
       },
       "original": {
@@ -80,18 +198,25 @@
         "type": "github"
       }
     },
-    "systems": {
+    "slimlock": {
+      "inputs": {
+        "nixpkgs": [
+          "dream2nix",
+          "purescript-overlay",
+          "nixpkgs"
+        ]
+      },
       "locked": {
-        "lastModified": 1681028828,
-        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
-        "owner": "nix-systems",
-        "repo": "default",
-        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "lastModified": 1688756706,
+        "narHash": "sha256-xzkkMv3neJJJ89zo3o2ojp7nFeaZc2G0fYwNXNJRFlo=",
+        "owner": "thomashoneyman",
+        "repo": "slimlock",
+        "rev": "cf72723f59e2340d24881fd7bf61cb113b4c407c",
         "type": "github"
       },
       "original": {
-        "owner": "nix-systems",
-        "repo": "default",
+        "owner": "thomashoneyman",
+        "repo": "slimlock",
         "type": "github"
       }
     },
@@ -102,11 +227,11 @@
         ]
       },
       "locked": {
-        "lastModified": 1762938485,
-        "narHash": "sha256-AlEObg0syDl+Spi4LsZIBrjw+snSVU4T8MOeuZJUJjM=",
+        "lastModified": 1768158989,
+        "narHash": "sha256-67vyT1+xClLldnumAzCTBvU0jLZ1YBcf4vANRWP3+Ak=",
         "owner": "numtide",
         "repo": "treefmt-nix",
-        "rev": "5b4ee75aeefd1e2d5a1cc43cf6ba65eba75e83e4",
+        "rev": "e96d59dff5c0d7fddb9d113ba108f03c3ef99eca",
         "type": "github"
       },
       "original": {

flake.nix

@@ -3,129 +3,134 @@
 
   inputs = {
     nixpkgs.url = "github:NixOS/nixpkgs/nixos-unstable";
-    flake-utils.url = "github:numtide/flake-utils";
-    # Provides Rust dev-env integration:
+
+    flake-parts = {
+      url = "github:hercules-ci/flake-parts";
+      inputs.nixpkgs-lib.follows = "nixpkgs";
+    };
+
+    crane.url = "github:ipetkov/crane";
+
     fenix = {
       url = "github:nix-community/fenix";
       inputs.nixpkgs.follows = "nixpkgs";
     };
-    # Provides formatting infrastructure:
+
     treefmt-nix = {
       url = "github:numtide/treefmt-nix";
       inputs.nixpkgs.follows = "nixpkgs";
     };
+
+    dream2nix = {
+      url = "github:nix-community/dream2nix";
+      inputs.nixpkgs.follows = "nixpkgs";
+    };
+
+    # Pinned nixpkgs for swift-format (swift is broken on x86_64-linux in newer nixpkgs)
+    nixpkgs-swift.url = "github:NixOS/nixpkgs/08dacfca559e1d7da38f3cf05f1f45ee9bfd213c";
   };
 
-  # TODO: figure out caching story
-  # nixConfig = {
-  #   # nix community cachix
-  #   extra-trusted-public-keys = "nix-community.cachix.org-1:mB9FSh9qf2dCimDSUo8Zy7bkq5CX+/rkCWyvRCYg3Fs=";
-  #   extra-substituters = "https://nix-community.cachix.org";
-  # };
+  nixConfig = {
+    extra-trusted-public-keys = "exo.cachix.org-1:okq7hl624TBeAR3kV+g39dUFSiaZgLRkLsFBCuJ2NZI=";
+    extra-substituters = "https://exo.cachix.org";
+  };
 
   outputs =
     inputs:
-    let
+    inputs.flake-parts.lib.mkFlake { inherit inputs; } {
       systems = [
         "x86_64-linux"
         "aarch64-darwin"
         "aarch64-linux"
       ];
-      fenixToolchain = system: inputs.fenix.packages.${system}.complete;
-    in
-    inputs.flake-utils.lib.eachSystem systems (
-      system:
-      let
-        pkgs = import inputs.nixpkgs {
-          inherit system;
-          overlays = [ inputs.fenix.overlays.default ];
-        };
-        treefmtEval = inputs.treefmt-nix.lib.evalModule pkgs {
-          projectRootFile = "flake.nix";
-          programs = {
-            nixpkgs-fmt.enable = true;
-            ruff-format = {
-              enable = true;
-              excludes = [ "rust/exo_pyo3_bindings/exo_pyo3_bindings.pyi" ];
+
+      imports = [
+        inputs.treefmt-nix.flakeModule
+        ./dashboard/parts.nix
+        ./rust/parts.nix
+      ];
+
+      perSystem =
+        { config, self', inputs', pkgs, lib, system, ... }:
+        let
+          fenixToolchain = inputs'.fenix.packages.complete;
+          # Use pinned nixpkgs for swift-format (swift is broken on x86_64-linux in newer nixpkgs)
+          pkgsSwift = import inputs.nixpkgs-swift { inherit system; };
+        in
+        {
+          treefmt = {
+            projectRootFile = "flake.nix";
+            programs = {
+              nixpkgs-fmt.enable = true;
+              ruff-format = {
+                enable = true;
+                excludes = [ "rust/exo_pyo3_bindings/exo_pyo3_bindings.pyi" ];
+              };
+              rustfmt = {
+                enable = true;
+                package = config.rust.toolchain;
+              };
+              prettier = {
+                enable = true;
+                includes = [ "*.ts" ];
+              };
+              swift-format = {
+                enable = true;
+                package = pkgsSwift.swiftPackages.swift-format;
+              };
             };
-            rustfmt = {
-              enable = true;
-              package = (fenixToolchain system).rustfmt;
-            };
-            prettier = {
-              enable = true;
-              includes = [ "*.ts" ];
-            };
-            swift-format.enable = true;
           };
-        };
-      in
-      {
-        formatter = treefmtEval.config.build.wrapper;
-        checks.formatting = treefmtEval.config.build.check inputs.self;
-        checks.lint = pkgs.runCommand "lint-check" { } ''
-          export RUFF_CACHE_DIR="$TMPDIR/ruff-cache"
-          ${pkgs.ruff}/bin/ruff check ${inputs.self}/
-          touch $out
-        '';
 
-        devShells.default = pkgs.mkShell {
-          packages =
-            with pkgs;
-            [
-              # PYTHON
-              python313
-              uv
-              ruff
-              basedpyright
-
-              # RUST
-              ((fenixToolchain system).withComponents [
-                "cargo"
-                "rustc"
-                "clippy"
-                "rustfmt"
-                "rust-src"
-              ])
-              rustup # Just here to make RustRover happy
-
-              # NIX
-              nixpkgs-fmt
-
-              # SVELTE
-              nodejs
-
-              # MISC
-              just
-              jq
-            ]
-            ++ (pkgs.lib.optionals pkgs.stdenv.isLinux [
-              # IFCONFIG
-              unixtools.ifconfig
-
-              # Build dependencies for Linux
-              pkg-config
-              openssl
-            ])
-            ++ (pkgs.lib.optionals pkgs.stdenv.isDarwin [
-              # MACMON
-              macmon
-            ]);
-
-          shellHook = ''
-            # PYTHON
-            export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:${pkgs.python313}/lib"
-            ${pkgs.lib.optionalString pkgs.stdenv.isLinux ''
-              # Build environment for Linux
-              export PKG_CONFIG_PATH="${pkgs.openssl.dev}/lib/pkgconfig:$PKG_CONFIG_PATH"
-              export LD_LIBRARY_PATH="${pkgs.openssl.out}/lib:$LD_LIBRARY_PATH"
-            ''}
-            echo
-            echo "🍎🍎 Run 'just <recipe>' to get started"
-            just --list
+          checks.lint = pkgs.runCommand "lint-check" { } ''
+            export RUFF_CACHE_DIR="$TMPDIR/ruff-cache"
+            ${pkgs.ruff}/bin/ruff check ${inputs.self}/
+            touch $out
           '';
 
+          devShells.default = with pkgs; pkgs.mkShell {
+            inputsFrom = [ self'.checks.cargo-build ];
+
+            packages =
+              [
+                # FORMATTING
+                config.treefmt.build.wrapper
+
+                # PYTHON
+                python313
+                uv
+                ruff
+                basedpyright
+
+                # RUST
+                config.rust.toolchain
+                maturin
+
+                # NIX
+                nixpkgs-fmt
+
+                # SVELTE
+                nodejs
+
+                # MISC
+                just
+                jq
+              ]
+              ++ lib.optionals stdenv.isLinux [
+                unixtools.ifconfig
+              ]
+              ++ lib.optionals stdenv.isDarwin [
+                macmon
+              ];
+
+            OPENSSL_NO_VENDOR = "1";
+
+            shellHook = ''
+              export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:${python313}/lib"
+              ${lib.optionalString stdenv.isLinux ''
+                export LD_LIBRARY_PATH="${openssl.out}/lib:$LD_LIBRARY_PATH"
+              ''}
+            '';
+          };
         };
-      }
-    );
+    };
 }

justfile

@@ -1,3 +1,5 @@
+export NIX_CONFIG := "extra-experimental-features = nix-command flakes"
+
 fmt:
     nix fmt
 

pyproject.toml

@@ -8,33 +8,22 @@ dependencies = [
     "aiofiles>=24.1.0",
     "aiohttp>=3.12.14",
     "types-aiofiles>=24.1.0.20250708",
-    "typeguard>=4.4.4",
     "pydantic>=2.11.7",
-    "base58>=2.1.1",
-    "cryptography>=45.0.5",
     "fastapi>=0.116.1",
     "filelock>=3.18.0",
-    "aiosqlite>=0.21.0",
-    "networkx>=3.5",
-    "protobuf>=6.32.0",
-    "rich>=14.1.0",
     "rustworkx>=0.17.1",
-    "sqlmodel>=0.0.24",
-    "sqlalchemy[asyncio]>=2.0.43",
-    "greenlet>=3.2.4",
     "huggingface-hub>=0.33.4",
     "psutil>=7.0.0",
     "loguru>=0.7.3",
-    "textual>=5.3.0",
     "exo_pyo3_bindings", # rust bindings
     "anyio==4.11.0",
-    "bidict>=0.23.1",
-    "mlx>=0.30.1; sys_platform == 'darwin'",
-    "mlx[cpu]>=0.30.1; sys_platform == 'linux'",
-    "mlx-lm>=0.28.3",
+    "mlx==0.30.1; sys_platform == 'darwin'",
+    "mlx[cpu]==0.30.1; sys_platform == 'linux'",
+    "mlx-lm @ git+https://github.com/AlexCheema/mlx-lm.git@fix-transformers-5.0.0rc2",
     "tiktoken>=0.12.0", # required for kimi k2 tokenizer
     "hypercorn>=0.18.0",
     "openai-harmony>=0.0.8",
+    "httpx>=0.28.1",
 ]
 
 [project.scripts]
@@ -45,6 +34,7 @@ exo = "exo.main:main"
 # dependencies only required for development
 [dependency-groups]
 dev = [
+    "basedpyright>=1.29.0",
     "pyinstaller>=6.17.0",
     "pytest>=8.4.0",
     "pytest-asyncio>=1.0.0",
@@ -110,6 +100,7 @@ root = "src"
 
 # supported platforms for this project
 [tool.uv]
+prerelease = "allow"
 environments = [
     "sys_platform == 'darwin'",
     "sys_platform == 'linux'",
@@ -135,3 +126,6 @@ env = [
   "EXO_TESTS=1"
 ]
 addopts = "-m 'not slow'"
+filterwarnings = [
+    "ignore:builtin type Swig:DeprecationWarning",
+]

rust/parts.nix

@@ -0,0 +1,145 @@
+{ inputs, ... }:
+{
+  perSystem =
+    { config, self', inputs', pkgs, lib, ... }:
+    let
+      # Fenix nightly toolchain with all components
+      fenixPkgs = inputs'.fenix.packages;
+      rustToolchain = fenixPkgs.complete.withComponents [
+        "cargo"
+        "rustc"
+        "clippy"
+        "rustfmt"
+        "rust-src"
+        "rust-analyzer"
+      ];
+
+      # Crane with fenix toolchain
+      craneLib = (inputs.crane.mkLib pkgs).overrideToolchain rustToolchain;
+
+      # Source filtering - only include rust/ directory and root Cargo files
+      # This ensures changes to Python/docs/etc don't trigger Rust rebuilds
+      src = lib.cleanSourceWith {
+        src = inputs.self;
+        filter =
+          path: type:
+          let
+            baseName = builtins.baseNameOf path;
+            parentDir = builtins.dirOf path;
+            inRustDir =
+              (lib.hasInfix "/rust/" path)
+              || (lib.hasSuffix "/rust" parentDir)
+              || (baseName == "rust" && type == "directory");
+            isRootCargoFile =
+              (baseName == "Cargo.toml" || baseName == "Cargo.lock")
+              && (builtins.dirOf path == toString inputs.self);
+          in
+          isRootCargoFile
+          || (inRustDir && (craneLib.filterCargoSources path type || lib.hasSuffix ".toml" path || lib.hasSuffix ".md" path));
+      };
+
+      # Common arguments for all Rust builds
+      commonArgs = {
+        inherit src;
+        pname = "exo-rust";
+        version = "0.0.1";
+        strictDeps = true;
+
+        nativeBuildInputs = [
+          pkgs.pkg-config
+          pkgs.python313 # Required for pyo3-build-config
+        ];
+
+        buildInputs = [
+          pkgs.openssl
+          pkgs.python313 # Required for pyo3 tests
+        ];
+
+        OPENSSL_NO_VENDOR = "1";
+
+        # Required for pyo3 tests to find libpython
+        LD_LIBRARY_PATH = lib.makeLibraryPath [ pkgs.python313 ];
+      };
+
+      # Build dependencies once for caching
+      cargoArtifacts = craneLib.buildDepsOnly (
+        commonArgs
+        // {
+          cargoExtraArgs = "--workspace";
+        }
+      );
+    in
+    {
+      # Export toolchain for use in treefmt and devShell
+      options.rust = {
+        toolchain = lib.mkOption {
+          type = lib.types.package;
+          default = rustToolchain;
+          description = "The Rust toolchain to use";
+        };
+      };
+
+      config = {
+        packages = {
+          # Python bindings wheel via maturin
+          exo_pyo3_bindings = craneLib.buildPackage (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              pname = "exo_pyo3_bindings";
+
+              nativeBuildInputs = commonArgs.nativeBuildInputs ++ [
+                pkgs.maturin
+              ];
+
+              buildPhaseCargoCommand = ''
+                maturin build \
+                  --release \
+                  --manylinux off \
+                  --manifest-path rust/exo_pyo3_bindings/Cargo.toml \
+                  --features "pyo3/extension-module,pyo3/experimental-async" \
+                  --interpreter ${pkgs.python313}/bin/python \
+                  --out dist
+              '';
+
+              # Don't use crane's default install behavior
+              doNotPostBuildInstallCargoBinaries = true;
+
+              installPhaseCommand = ''
+                mkdir -p $out
+                cp dist/*.whl $out/
+              '';
+            }
+          );
+        };
+
+        checks = {
+          # Full workspace build (all crates)
+          cargo-build = craneLib.buildPackage (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              cargoExtraArgs = "--workspace";
+            }
+          );
+          # Run tests with nextest
+          cargo-nextest = craneLib.cargoNextest (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              cargoExtraArgs = "--workspace";
+            }
+          );
+
+          # Build documentation
+          cargo-doc = craneLib.cargoDoc (
+            commonArgs
+            // {
+              inherit cargoArtifacts;
+              cargoExtraArgs = "--workspace";
+            }
+          );
+        };
+      };
+    };
+}

rust/system_custodian/Cargo.toml

@@ -1,47 +0,0 @@
-[package]
-name = "system_custodian"
-version = { workspace = true }
-edition = { workspace = true }
-publish = false
-
-[lib]
-doctest = false
-name = "system_custodian"
-path = "src/lib.rs"
-
-[[bin]]
-path = "src/bin/main.rs"
-name = "system_custodian"
-doc = false
-
-[lints]
-workspace = true
-
-[dependencies]
-# datastructures
-either = { workspace = true }
-
-# macro dependencies
-extend = { workspace = true }
-delegate = { workspace = true }
-impl-trait-for-tuples = { workspace = true }
-derive_more = { workspace = true }
-
-# async
-tokio = { workspace = true, features = ["full"] }
-futures = { workspace = true }
-futures-timer = { workspace = true }
-
-# utility dependencies
-util = { workspace = true }
-thiserror = { workspace = true }
-#internment = { workspace = true }
-#recursion = { workspace = true }
-#generativity = { workspace = true }
-#itertools = { workspace = true }
-tracing-subscriber = { version = "0.3.19", features = ["default", "env-filter"] }
-keccak-const = { workspace = true }
-
-# tracing/logging
-log = { workspace = true }
-

rust/system_custodian/src/bin/main.rs

@@ -1,4 +0,0 @@
-//! TODO: documentation
-//!
-
-fn main() {}

rust/system_custodian/src/lib.rs

@@ -1,69 +0,0 @@
-//! This crate defines the logic of, and ways to interact with, Exo's **_System Custodian_** daemon.
-//!
-//! The **_System Custodian_** daemon is supposed to be a long-living process that precedes the
-//! launch of the Exo application, and responsible for ensuring the system (configuration, settings,
-//! etc.) is in an appropriate state to facilitate the running of Exo application.
-//! The **_System Custodian_** daemon shall expose a [D-Bus](https://www.freedesktop.org/wiki/Software/dbus/)
-//! service which Exo application use to _control & query_ it.
-//!
-//! # Lifecycle
-//! When the Exo application starts, it will _wake_ the **_System Custodian_** daemon for the
-//! duration of its lifetime, and after it has terminated the daemon will go back to sleep. When
-//! the daemon wakes up, it will configure the system into a state suitable for the Exo Application;
-//! When the daemon goes to sleep, it will revert those changes as much as it can in case they were
-//! destructive to the user's pre-existing configurations.
-//!
-//! # Responsibilities
-//! TODO: these are purely on MacOS, but change to be more broad
-//! The **_System Custodian_** daemon is responsible for using System Configuration framework to
-//!  1. duplicate the current network set
-//!  2. modify existing services to turn on IPv6 if not there
-//!  3. remove any bridge services & add any missing services that AREN'T bridge
-//! TODO: In the future:
-//!  1. run a dummy AWDL service to [allow for macOS peer-to-peer wireless networking](https://yggdrasil-network.github.io/2019/08/19/awdl.html)
-//!  2. toggle some GPU/memory configurations to speed up GPU (ask Alex what those configurations are)
-//!  3. if we ever decide to provide our **own network interfaces** that abstract over some userland
-//!     logic, this would be the place to spin that up.
-//!
-//! Then it will watch the SCDynamicStore for:
-//!  1. all __actual__ network interfaces -> collect information on them e.g. their BSD name, MAC
-//!     address, MTU, IPv6 addresses, etc. -> and set up watchers/notifiers to inform the DBus
-//!     interface of any changes
-//!  2. watch for any __undesirable__ changes to configuration and revert it
-//!
-//! It should somehow (probably through system sockets and/or BSD interface) trigger IPv6 NDP on
-//! each of the interfaces & also listen to/query for any changes on the OS routing cache??
-//! Basically emulate the `ping6 ff02::1%enX` and `ndp -an` commands BUT BETTER!!!
-//!  1. all that info should coalesce back to the overall state colleted -> should be queryable
-//!     over D-Bus
-//! TODO:
-//!  1. we might potentially add to this step a handshake of some kind...? To ensure that we can
-//!     ACTUALLY communicate with that machine over that link over e.g. TCP, UDP, etc. Will the
-//!     handshake require to know Node ID? Will the handshake require heartbeats? Who knows...
-//!  2. if we ever decide to write proprietary L2/L3 protocols for quicker communication,
-//!     e.g. [AF_NDRV](https://www.zerotier.com/blog/how-zerotier-eliminated-kernel-extensions-on-macos/)
-//!     for raw ethernet frame communication, or even a [custom thunderbolt PCIe driver](https://developer.apple.com/documentation/pcidriverkit/creating-custom-pcie-drivers-for-thunderbolt-devices),
-//!     then this would be the place to carry out discovery and propper handshakes with devices
-//!     on the other end of the link.
-//!
-
-// enable Rust-unstable features for convenience
-#![feature(trait_alias)]
-#![feature(stmt_expr_attributes)]
-#![feature(type_alias_impl_trait)]
-#![feature(specialization)]
-#![feature(unboxed_closures)]
-#![feature(const_trait_impl)]
-#![feature(fn_traits)]
-
-pub(crate) mod private {
-    // sealed traits support
-    pub trait Sealed {}
-    impl<T: ?Sized> Sealed for T {}
-}
-
-/// Namespace for all the type/trait aliases used by this crate.
-pub(crate) mod alias {}
-
-/// Namespace for crate-wide extension traits/methods
-pub(crate) mod ext {}

src/exo/main.py

@@ -1,6 +1,7 @@
 import argparse
 import multiprocessing as mp
 import os
+import resource
 import signal
 from dataclasses import dataclass, field
 from typing import Self
@@ -195,6 +196,8 @@ class Node:
 
 def main():
     args = Args.parse()
+    soft, hard = resource.getrlimit(resource.RLIMIT_NOFILE)
+    resource.setrlimit(resource.RLIMIT_NOFILE, (max(soft, 65535), hard))
 
     mp.set_start_method("spawn")
     # TODO: Refactor the current verbosity system
@@ -202,6 +205,14 @@ def main():
     logger.info("Starting EXO")
     logger.info(f"EXO_LIBP2P_NAMESPACE: {os.getenv('EXO_LIBP2P_NAMESPACE')}")
 
+    # Set FAST_SYNCH override env var for runner subprocesses
+    if args.fast_synch is True:
+        os.environ["EXO_FAST_SYNCH"] = "on"
+        logger.info("FAST_SYNCH forced ON")
+    elif args.fast_synch is False:
+        os.environ["EXO_FAST_SYNCH"] = "off"
+        logger.info("FAST_SYNCH forced OFF")
+
     node = anyio.run(Node.create, args)
     anyio.run(node.run)
     logger.info("EXO Shutdown complete")
@@ -215,6 +226,7 @@ class Args(CamelCaseModel):
     api_port: PositiveInt = 52415
     tb_only: bool = False
     no_worker: bool = False
+    fast_synch: bool | None = None  # None = auto, True = force on, False = force off
 
     @classmethod
     def parse(cls) -> Self:
@@ -256,6 +268,20 @@ class Args(CamelCaseModel):
             "--no-worker",
             action="store_true",
         )
+        fast_synch_group = parser.add_mutually_exclusive_group()
+        fast_synch_group.add_argument(
+            "--fast-synch",
+            action="store_true",
+            dest="fast_synch",
+            default=None,
+            help="Force MLX FAST_SYNCH on (for JACCL backend)",
+        )
+        fast_synch_group.add_argument(
+            "--no-fast-synch",
+            action="store_false",
+            dest="fast_synch",
+            help="Force MLX FAST_SYNCH off",
+        )
 
         args = parser.parse_args()
         return cls(**vars(args))  # pyright: ignore[reportAny] - We are intentionally validating here, we can't do it statically

src/exo/master/api.py

@@ -1,24 +1,19 @@
 import time
 from collections.abc import AsyncGenerator
+from http import HTTPStatus
 from typing import cast
 
 import anyio
-from anyio import create_task_group
+from anyio import BrokenResourceError, create_task_group
 from anyio.abc import TaskGroup
-from fastapi import FastAPI, HTTPException
+from fastapi import FastAPI, HTTPException, Request
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import StreamingResponse
+from fastapi.responses import JSONResponse, StreamingResponse
 from fastapi.staticfiles import StaticFiles
 from hypercorn.asyncio import serve  # pyright: ignore[reportUnknownVariableType]
 from hypercorn.config import Config
 from hypercorn.typing import ASGIFramework
 from loguru import logger
-from openai_harmony import (  # pyright: ignore[reportMissingTypeStubs]
-    HarmonyEncodingName,
-    Role,
-    StreamableParser,
-    load_harmony_encoding,
-)
 
 from exo.master.placement import place_instance as get_instance_placements
 from exo.shared.apply import apply
@@ -35,6 +30,8 @@ from exo.shared.types.api import (
     CreateInstanceParams,
     CreateInstanceResponse,
     DeleteInstanceResponse,
+    ErrorInfo,
+    ErrorResponse,
     FinishReason,
     GenerationStats,
     ModelList,
@@ -55,7 +52,12 @@ from exo.shared.types.commands import (
     TaskFinished,
 )
 from exo.shared.types.common import CommandId, NodeId, SessionId
-from exo.shared.types.events import ChunkGenerated, Event, ForwarderEvent, IndexedEvent
+from exo.shared.types.events import (
+    ChunkGenerated,
+    Event,
+    ForwarderEvent,
+    IndexedEvent,
+)
 from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelId, ModelMetadata
 from exo.shared.types.state import State
@@ -67,8 +69,6 @@ from exo.utils.channels import Receiver, Sender, channel
 from exo.utils.dashboard_path import find_dashboard
 from exo.utils.event_buffer import OrderedBuffer
 
-encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)
-
 
 def chunk_to_response(
     chunk: TokenChunk, command_id: CommandId
@@ -123,6 +123,7 @@ class API:
         self.paused_ev: anyio.Event = anyio.Event()
 
         self.app = FastAPI()
+        self._setup_exception_handlers()
         self._setup_cors()
         self._setup_routes()
 
@@ -153,6 +154,21 @@ class API:
         self.paused_ev.set()
         self.paused_ev = anyio.Event()
 
+    def _setup_exception_handlers(self) -> None:
+        self.app.exception_handler(HTTPException)(self.http_exception_handler)
+
+    async def http_exception_handler(
+        self, _: Request, exc: HTTPException
+    ) -> JSONResponse:
+        err = ErrorResponse(
+            error=ErrorInfo(
+                message=exc.detail,
+                type=HTTPStatus(exc.status_code).phrase,
+                code=exc.status_code,
+            )
+        )
+        return JSONResponse(err.model_dump(), status_code=exc.status_code)
+
     def _setup_cors(self) -> None:
         self.app.add_middleware(
             CORSMiddleware,
@@ -236,6 +252,7 @@ class API:
                     instance_meta=instance_meta,
                     min_nodes=min_nodes,
                 ),
+                node_profiles=self.state.node_profiles,
                 topology=self.state.topology,
                 current_instances=self.state.instances,
             )
@@ -291,6 +308,7 @@ class API:
                             instance_meta=instance_meta,
                             min_nodes=min_nodes,
                         ),
+                        node_profiles=self.state.node_profiles,
                         topology=self.state.topology,
                         current_instances=self.state.instances,
                     )
@@ -381,35 +399,8 @@ class API:
             instance_id=instance_id,
         )
 
-    async def _process_gpt_oss(self, token_chunks: Receiver[TokenChunk]):
-        stream = StreamableParser(encoding, role=Role.ASSISTANT)
-        thinking = False
-
-        async for chunk in token_chunks:
-            stream.process(chunk.token_id)
-
-            delta = stream.last_content_delta
-            ch = stream.current_channel
-
-            if ch == "analysis" and not thinking:
-                thinking = True
-                yield chunk.model_copy(update={"text": "<think>"})
-
-            if ch != "analysis" and thinking:
-                thinking = False
-                yield chunk.model_copy(update={"text": "</think>"})
-
-            if delta:
-                yield chunk.model_copy(update={"text": delta})
-
-            if chunk.finish_reason is not None:
-                if thinking:
-                    yield chunk.model_copy(update={"text": "</think>"})
-                yield chunk
-                break
-
     async def _chat_chunk_stream(
-        self, command_id: CommandId, parse_gpt_oss: bool
+        self, command_id: CommandId
     ) -> AsyncGenerator[TokenChunk, None]:
         """Yield `TokenChunk`s for a given command until completion."""
 
@@ -417,16 +408,10 @@ class API:
             self._chat_completion_queues[command_id], recv = channel[TokenChunk]()
 
             with recv as token_chunks:
-                if parse_gpt_oss:
-                    async for chunk in self._process_gpt_oss(token_chunks):
-                        yield chunk
-                        if chunk.finish_reason is not None:
-                            break
-                else:
-                    async for chunk in token_chunks:
-                        yield chunk
-                        if chunk.finish_reason is not None:
-                            break
+                async for chunk in token_chunks:
+                    yield chunk
+                    if chunk.finish_reason is not None:
+                        break
 
         except anyio.get_cancelled_exc_class():
             # TODO: TaskCancelled
@@ -442,11 +427,23 @@ class API:
             del self._chat_completion_queues[command_id]
 
     async def _generate_chat_stream(
-        self, command_id: CommandId, parse_gpt_oss: bool
+        self, command_id: CommandId
     ) -> AsyncGenerator[str, None]:
         """Generate chat completion stream as JSON strings."""
 
-        async for chunk in self._chat_chunk_stream(command_id, parse_gpt_oss):
+        async for chunk in self._chat_chunk_stream(command_id):
+            if chunk.finish_reason == "error":
+                error_response = ErrorResponse(
+                    error=ErrorInfo(
+                        message=chunk.error_message or "Internal server error",
+                        type="InternalServerError",
+                        code=500,
+                    )
+                )
+                yield f"data: {error_response.model_dump_json()}\n\n"
+                yield "data: [DONE]\n\n"
+                return
+
             chunk_response: ChatCompletionResponse = chunk_to_response(
                 chunk, command_id
             )
@@ -458,7 +455,7 @@ class API:
                 yield "data: [DONE]\n\n"
 
     async def _collect_chat_completion(
-        self, command_id: CommandId, parse_gpt_oss: bool
+        self, command_id: CommandId
     ) -> ChatCompletionResponse:
         """Collect all token chunks for a chat completion and return a single response."""
 
@@ -466,7 +463,13 @@ class API:
         model: str | None = None
         finish_reason: FinishReason | None = None
 
-        async for chunk in self._chat_chunk_stream(command_id, parse_gpt_oss):
+        async for chunk in self._chat_chunk_stream(command_id):
+            if chunk.finish_reason == "error":
+                raise HTTPException(
+                    status_code=500,
+                    detail=chunk.error_message or "Internal server error",
+                )
+
             if model is None:
                 model = chunk.model
 
@@ -495,7 +498,7 @@ class API:
         )
 
     async def _collect_chat_completion_with_stats(
-        self, command_id: CommandId, parse_gpt_oss: bool
+        self, command_id: CommandId
     ) -> BenchChatCompletionResponse:
         text_parts: list[str] = []
         model: str | None = None
@@ -503,7 +506,13 @@ class API:
 
         stats: GenerationStats | None = None
 
-        async for chunk in self._chat_chunk_stream(command_id, parse_gpt_oss):
+        async for chunk in self._chat_chunk_stream(command_id):
+            if chunk.finish_reason == "error":
+                raise HTTPException(
+                    status_code=500,
+                    detail=chunk.error_message or "Internal server error",
+                )
+
             if model is None:
                 model = chunk.model
 
@@ -544,8 +553,6 @@ class API:
         """Handle chat completions, supporting both streaming and non-streaming responses."""
         model_meta = await resolve_model_meta(payload.model)
         payload.model = model_meta.model_id
-        parse_gpt_oss = "gpt-oss" in model_meta.model_id.lower()
-        logger.info(f"{parse_gpt_oss=}")
 
         if not any(
             instance.shard_assignments.model_id == payload.model
@@ -562,17 +569,16 @@ class API:
         await self._send(command)
         if payload.stream:
             return StreamingResponse(
-                self._generate_chat_stream(command.command_id, parse_gpt_oss),
+                self._generate_chat_stream(command.command_id),
                 media_type="text/event-stream",
             )
 
-        return await self._collect_chat_completion(command.command_id, parse_gpt_oss)
+        return await self._collect_chat_completion(command.command_id)
 
     async def bench_chat_completions(
         self, payload: BenchChatCompletionTaskParams
     ) -> BenchChatCompletionResponse:
         model_meta = await resolve_model_meta(payload.model)
-        parse_gpt_oss = "gpt-oss" in model_meta.model_id.lower()
         payload.model = model_meta.model_id
 
         if not any(
@@ -589,19 +595,15 @@ class API:
         command = ChatCompletion(request_params=payload)
         await self._send(command)
 
-        response = await self._collect_chat_completion_with_stats(
-            command.command_id,
-            parse_gpt_oss,
-        )
+        response = await self._collect_chat_completion_with_stats(command.command_id)
         return response
 
     def _calculate_total_available_memory(self) -> Memory:
         """Calculate total available memory across all nodes in bytes."""
         total_available = Memory()
 
-        for node in self.state.topology.list_nodes():
-            if node.node_profile is not None:
-                total_available += node.node_profile.memory.ram_available
+        for profile in self.state.node_profiles.values():
+            total_available += profile.memory.ram_available
 
         return total_available
 
@@ -654,14 +656,14 @@ class API:
                 for idx, event in self.event_buffer.drain_indexed():
                     self._event_log.append(event)
                     self.state = apply(self.state, IndexedEvent(event=event, idx=idx))
-                    if (
-                        isinstance(event, ChunkGenerated)
-                        and event.command_id in self._chat_completion_queues
-                    ):
+                    if isinstance(event, ChunkGenerated):
                         assert isinstance(event.chunk, TokenChunk)
-                        await self._chat_completion_queues[event.command_id].send(
-                            event.chunk
-                        )
+                        queue = self._chat_completion_queues.get(event.command_id)
+                        if queue is not None:
+                            try:
+                                await queue.send(event.chunk)
+                            except BrokenResourceError:
+                                self._chat_completion_queues.pop(event.command_id, None)
 
     async def _pause_on_new_election(self):
         with self.election_receiver as ems:

src/exo/master/main.py

@@ -27,6 +27,7 @@ from exo.shared.types.events import (
     ForwarderEvent,
     IndexedEvent,
     InstanceDeleted,
+    NodeGatheredInfo,
     NodeTimedOut,
     TaskCreated,
     TaskDeleted,
@@ -158,6 +159,7 @@ class Master:
                                 command,
                                 self.state.topology,
                                 self.state.instances,
+                                self.state.node_profiles,
                             )
                             transition_events = get_transition_events(
                                 self.state.instances, placement
@@ -200,9 +202,7 @@ class Master:
     async def _plan(self) -> None:
         while True:
             # kill broken instances
-            connected_node_ids = set(
-                [x.node_id for x in self.state.topology.list_nodes()]
-            )
+            connected_node_ids = set(self.state.topology.list_nodes())
             for instance_id, instance in self.state.instances.items():
                 for node_id in instance.shard_assignments.node_to_runner:
                     if node_id not in connected_node_ids:
@@ -237,6 +237,8 @@ class Master:
                     self.state = apply(self.state, indexed)
 
                     event._master_time_stamp = datetime.now(tz=timezone.utc)  # pyright: ignore[reportPrivateUsage]
+                    if isinstance(event, NodeGatheredInfo):
+                        event.when = str(datetime.now(tz=timezone.utc))
 
                     self._event_log.append(event)
                     await self._send_event(indexed)

src/exo/master/placement.py

@@ -6,9 +6,10 @@ from typing import Sequence
 from loguru import logger
 
 from exo.master.placement_utils import (
+    Cycle,
     filter_cycles_by_memory,
-    get_mlx_ibv_devices_matrix,
     get_mlx_jaccl_coordinators,
+    get_mlx_jaccl_devices_matrix,
     get_mlx_ring_hosts_by_node,
     get_shard_assignments,
     get_smallest_cycles,
@@ -19,10 +20,11 @@ from exo.shared.types.commands import (
     DeleteInstance,
     PlaceInstance,
 )
+from exo.shared.types.common import NodeId
 from exo.shared.types.events import Event, InstanceCreated, InstanceDeleted
 from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelId
-from exo.shared.types.topology import NodeInfo
+from exo.shared.types.profiling import NodePerformanceProfile
 from exo.shared.types.worker.instances import (
     Instance,
     InstanceId,
@@ -52,19 +54,14 @@ def place_instance(
     command: PlaceInstance,
     topology: Topology,
     current_instances: Mapping[InstanceId, Instance],
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
 ) -> dict[InstanceId, Instance]:
-    all_nodes = list(topology.list_nodes())
-
-    logger.info("finding cycles:")
     cycles = topology.get_cycles()
-    singleton_cycles = [[node] for node in all_nodes]
-    candidate_cycles = list(
-        filter(lambda it: len(it) >= command.min_nodes, cycles + singleton_cycles)
-    )
+    candidate_cycles = list(filter(lambda it: len(it) >= command.min_nodes, cycles))
     cycles_with_sufficient_memory = filter_cycles_by_memory(
-        candidate_cycles, command.model_meta.storage_size
+        candidate_cycles, node_profiles, command.model_meta.storage_size
     )
-    if not cycles_with_sufficient_memory:
+    if len(cycles_with_sufficient_memory) == 0:
         raise ValueError("No cycles found with sufficient memory")
 
     if command.sharding == Sharding.Tensor:
@@ -92,44 +89,38 @@ def place_instance(
     smallest_cycles = get_smallest_cycles(cycles_with_sufficient_memory)
 
     smallest_tb_cycles = [
-        cycle
-        for cycle in smallest_cycles
-        if topology.get_subgraph_from_nodes(cycle).is_thunderbolt_cycle(cycle)
+        cycle for cycle in smallest_cycles if topology.is_thunderbolt_cycle(cycle)
     ]
 
     if smallest_tb_cycles != []:
         smallest_cycles = smallest_tb_cycles
 
-    cycles_with_leaf_nodes: list[list[NodeInfo]] = [
+    cycles_with_leaf_nodes: list[Cycle] = [
         cycle
         for cycle in smallest_cycles
-        if any(topology.node_is_leaf(node.node_id) for node in cycle)
+        if any(topology.node_is_leaf(node_id) for node_id in cycle)
     ]
 
     selected_cycle = max(
         cycles_with_leaf_nodes if cycles_with_leaf_nodes != [] else smallest_cycles,
         key=lambda cycle: sum(
-            (
-                node.node_profile.memory.ram_available
-                for node in cycle
-                if node.node_profile is not None
-            ),
+            (node_profiles[node_id].memory.ram_available for node_id in cycle),
             start=Memory(),
         ),
     )
 
     shard_assignments = get_shard_assignments(
-        command.model_meta, selected_cycle, command.sharding
+        command.model_meta, selected_cycle, command.sharding, node_profiles
     )
 
-    cycle_digraph: Topology = topology.get_subgraph_from_nodes(selected_cycle)
+    cycle_digraph: Topology = topology.get_subgraph_from_nodes(selected_cycle.node_ids)
 
     instance_id = InstanceId()
     target_instances = dict(deepcopy(current_instances))
 
     if len(selected_cycle) == 1:
         logger.warning(
-            "You have likely selected ibv for a single node instance; falling back to MlxRing"
+            "You have likely selected jaccl for a single node instance; falling back to MlxRing"
         )
 
         command.instance_meta = InstanceMeta.MlxRing
@@ -137,19 +128,20 @@ def place_instance(
     # TODO: Single node instances
     match command.instance_meta:
         case InstanceMeta.MlxJaccl:
-            mlx_ibv_devices = get_mlx_ibv_devices_matrix(
-                selected_cycle,
+            mlx_jaccl_devices = get_mlx_jaccl_devices_matrix(
+                [node_id for node_id in selected_cycle],
                 cycle_digraph,
             )
             mlx_jaccl_coordinators = get_mlx_jaccl_coordinators(
-                selected_cycle,
+                coordinator=selected_cycle.node_ids[0],
                 coordinator_port=random_ephemeral_port(),
                 cycle_digraph=cycle_digraph,
+                node_profiles=node_profiles,
             )
             target_instances[instance_id] = MlxJacclInstance(
                 instance_id=instance_id,
                 shard_assignments=shard_assignments,
-                ibv_devices=mlx_ibv_devices,
+                jaccl_devices=mlx_jaccl_devices,
                 jaccl_coordinators=mlx_jaccl_coordinators,
             )
         case InstanceMeta.MlxRing:
@@ -158,6 +150,7 @@ def place_instance(
                 selected_cycle=selected_cycle,
                 cycle_digraph=cycle_digraph,
                 ephemeral_port=ephemeral_port,
+                node_profiles=node_profiles,
             )
             target_instances[instance_id] = MlxRingInstance(
                 instance_id=instance_id,

src/exo/master/placement_utils.py

@@ -1,15 +1,13 @@
-from collections.abc import Generator
-from typing import TypeGuard, cast
+from collections.abc import Generator, Mapping
 
 from loguru import logger
-from pydantic import BaseModel
 
 from exo.shared.topology import Topology
 from exo.shared.types.common import Host, NodeId
 from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelMetadata
 from exo.shared.types.profiling import NodePerformanceProfile
-from exo.shared.types.topology import NodeInfo
+from exo.shared.types.topology import Cycle, RDMAConnection, SocketConnection
 from exo.shared.types.worker.runners import RunnerId, ShardAssignments
 from exo.shared.types.worker.shards import (
     PipelineShardMetadata,
@@ -19,63 +17,110 @@ from exo.shared.types.worker.shards import (
 )
 
 
-class NodeWithProfile(BaseModel):
-    node_id: NodeId
-    node_profile: NodePerformanceProfile
-
-
-def narrow_all_nodes(nodes: list[NodeInfo]) -> TypeGuard[list[NodeWithProfile]]:
-    return all(node.node_profile is not None for node in nodes)
-
-
 def filter_cycles_by_memory(
-    cycles: list[list[NodeInfo]], required_memory: Memory
-) -> list[list[NodeInfo]]:
-    filtered_cycles: list[list[NodeInfo]] = []
+    cycles: list[Cycle],
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
+    required_memory: Memory,
+) -> list[Cycle]:
+    filtered_cycles: list[Cycle] = []
     for cycle in cycles:
-        if not narrow_all_nodes(cycle):
+        if not all(node in node_profiles for node in cycle):
             continue
 
         total_mem = sum(
-            (node.node_profile.memory.ram_available for node in cycle), start=Memory()
+            (node_profiles[node_id].memory.ram_available for node_id in cycle.node_ids),
+            start=Memory(),
         )
         if total_mem >= required_memory:
-            filtered_cycles.append(cast(list[NodeInfo], cycle))
+            filtered_cycles.append(cycle)
     return filtered_cycles
 
 
-def get_smallest_cycles(cycles: list[list[NodeInfo]]) -> list[list[NodeInfo]]:
+def get_smallest_cycles(
+    cycles: list[Cycle],
+) -> list[Cycle]:
     min_nodes = min(len(cycle) for cycle in cycles)
     return [cycle for cycle in cycles if len(cycle) == min_nodes]
 
 
+def allocate_layers_proportionally(
+    total_layers: int,
+    memory_fractions: list[float],
+) -> list[int]:
+    n = len(memory_fractions)
+    if n == 0:
+        raise ValueError("Cannot allocate layers to an empty node list")
+    if total_layers < n:
+        raise ValueError(
+            f"Cannot distribute {total_layers} layers across {n} nodes "
+            "(need at least 1 layer per node)"
+        )
+
+    # Largest remainder: floor each, then distribute remainder by fractional part
+    raw = [f * total_layers for f in memory_fractions]
+    result = [int(r) for r in raw]
+    by_remainder = sorted(range(n), key=lambda i: raw[i] - result[i], reverse=True)
+    for i in range(total_layers - sum(result)):
+        result[by_remainder[i]] += 1
+
+    # Ensure minimum 1 per node by taking from the largest
+    for i in range(n):
+        if result[i] == 0:
+            max_idx = max(range(n), key=lambda j: result[j])
+            assert result[max_idx] > 1
+            result[max_idx] -= 1
+            result[i] = 1
+
+    return result
+
+
 def get_shard_assignments_for_pipeline_parallel(
     model_meta: ModelMetadata,
-    selected_cycle: list[NodeWithProfile],
+    cycle: Cycle,
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
 ):
+    if not cycle.node_ids:
+        raise ValueError("Cannot create shard assignments for empty node cycle")
+
     cycle_memory = sum(
-        (node.node_profile.memory.ram_available for node in selected_cycle),
+        (node_profiles[node_id].memory.ram_available for node_id in cycle.node_ids),
         start=Memory(),
     )
+
+    if cycle_memory.in_bytes == 0:
+        raise ValueError("Cannot create shard assignments: total available memory is 0")
+
     total_layers = model_meta.n_layers
-    world_size = len(selected_cycle)
+    world_size = len(cycle)
     runner_to_shard: dict[RunnerId, ShardMetadata] = {}
     node_to_runner: dict[NodeId, RunnerId] = {}
 
-    layers_assigned = 0
-    for i, node in enumerate(selected_cycle):
-        if i == len(selected_cycle) - 1:
-            node_layers = total_layers - layers_assigned
-        else:
-            node_layers = round(
-                total_layers
-                * (
-                    node.node_profile.memory.ram_available.in_bytes
-                    / cycle_memory.in_bytes
-                )
-            )
-            node_layers = max(1, node_layers)
+    layer_allocations = allocate_layers_proportionally(
+        total_layers=total_layers,
+        memory_fractions=[
+            node_profiles[node_id].memory.ram_available.in_bytes / cycle_memory.in_bytes
+            for node_id in cycle.node_ids
+        ],
+    )
 
+    # Validate each node has sufficient memory for its assigned layers
+    memory_per_layer = model_meta.storage_size.in_bytes / total_layers
+    for i, (node_id, node_layers) in enumerate(
+        zip(cycle.node_ids, layer_allocations, strict=True)
+    ):
+        required_memory = node_layers * memory_per_layer
+        available_memory = node_profiles[node_id].memory.ram_available.in_bytes
+        if required_memory > available_memory:
+            raise ValueError(
+                f"Node {i} ({node_id}) has insufficient memory: "
+                f"requires {required_memory / (1024**3):.2f} GB for {node_layers} layers, "
+                f"but only has {available_memory / (1024**3):.2f} GB available"
+            )
+
+    layers_assigned = 0
+    for i, (node_id, node_layers) in enumerate(
+        zip(cycle.node_ids, layer_allocations, strict=True)
+    ):
         runner_id = RunnerId()
 
         shard = PipelineShardMetadata(
@@ -88,7 +133,7 @@ def get_shard_assignments_for_pipeline_parallel(
         )
 
         runner_to_shard[runner_id] = shard
-        node_to_runner[node.node_id] = runner_id
+        node_to_runner[node_id] = runner_id
         layers_assigned += node_layers
 
     shard_assignments = ShardAssignments(
@@ -102,14 +147,14 @@ def get_shard_assignments_for_pipeline_parallel(
 
 def get_shard_assignments_for_tensor_parallel(
     model_meta: ModelMetadata,
-    selected_cycle: list[NodeWithProfile],
+    cycle: Cycle,
 ):
     total_layers = model_meta.n_layers
-    world_size = len(selected_cycle)
+    world_size = len(cycle)
     runner_to_shard: dict[RunnerId, ShardMetadata] = {}
     node_to_runner: dict[NodeId, RunnerId] = {}
 
-    for i, node in enumerate(selected_cycle):
+    for i, node_id in enumerate(cycle):
         shard = TensorShardMetadata(
             model_meta=model_meta,
             device_rank=i,
@@ -122,7 +167,7 @@ def get_shard_assignments_for_tensor_parallel(
         runner_id = RunnerId()
 
         runner_to_shard[runner_id] = shard
-        node_to_runner[node.node_id] = runner_id
+        node_to_runner[node_id] = runner_id
 
     shard_assignments = ShardAssignments(
         model_id=model_meta.model_id,
@@ -135,21 +180,21 @@ def get_shard_assignments_for_tensor_parallel(
 
 def get_shard_assignments(
     model_meta: ModelMetadata,
-    selected_cycle: list[NodeInfo],
+    cycle: Cycle,
     sharding: Sharding,
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
 ) -> ShardAssignments:
-    if not narrow_all_nodes(selected_cycle):
-        raise ValueError("All nodes must have profiles to create shard assignments")
     match sharding:
         case Sharding.Pipeline:
             return get_shard_assignments_for_pipeline_parallel(
                 model_meta=model_meta,
-                selected_cycle=selected_cycle,
+                cycle=cycle,
+                node_profiles=node_profiles,
             )
         case Sharding.Tensor:
             return get_shard_assignments_for_tensor_parallel(
                 model_meta=model_meta,
-                selected_cycle=selected_cycle,
+                cycle=cycle,
             )
 
 
@@ -164,38 +209,40 @@ def get_hosts_from_subgraph(cycle_digraph: Topology) -> list[Host]:
             )
         return []
 
+    cycle = cycles[0]
+
     get_thunderbolt = False
-    if cycle_digraph.is_thunderbolt_cycle(cycles[0]):
+    if cycle_digraph.is_thunderbolt_cycle(cycle):
         get_thunderbolt = True
 
     logger.info(f"Using thunderbolt cycle: {get_thunderbolt}")
 
-    cycle = cycles[0]
     hosts: list[Host] = []
     for i in range(len(cycle)):
-        current_node = cycle[i]
-        next_node = cycle[(i + 1) % len(cycle)]
+        current_node = cycle.node_ids[i]
+        next_node = cycle.node_ids[(i + 1) % len(cycle)]
 
-        for connection in cycle_digraph.list_connections():
-            if (
-                connection.local_node_id == current_node.node_id
-                and connection.send_back_node_id == next_node.node_id
-            ):
-                if get_thunderbolt and not connection.is_thunderbolt():
-                    continue
-                assert connection.send_back_multiaddr is not None
-                host = Host(
-                    ip=connection.send_back_multiaddr.ip_address,
-                    port=connection.send_back_multiaddr.port,
-                )
-                hosts.append(host)
-                break
+        for connection in cycle_digraph.get_all_connections_between(
+            source=current_node, sink=next_node
+        ):
+            if not isinstance(connection, SocketConnection):
+                continue
+
+            if get_thunderbolt and not connection.is_thunderbolt():
+                continue
+
+            host = Host(
+                ip=connection.sink_multiaddr.ip_address,
+                port=connection.sink_multiaddr.port,
+            )
+            hosts.append(host)
+            break
 
     return hosts
 
 
-def get_mlx_ibv_devices_matrix(
-    selected_cycle: list[NodeInfo],
+def get_mlx_jaccl_devices_matrix(
+    selected_cycle: list[NodeId],
     cycle_digraph: Topology,
 ) -> list[list[str | None]]:
     """Build connectivity matrix mapping device i to device j via RDMA interface names.
@@ -214,72 +261,37 @@ def get_mlx_ibv_devices_matrix(
             if i == j:
                 continue
 
-            # Find the IP J uses to talk to I
-            for connection_ip, _ in _find_connection_ip(node_j, node_i, cycle_digraph):
-                # This is a local IP on I, which is attached to an interface: find that interface
-                if interface_name := _find_rdma_interface_name_for_ip(
-                    connection_ip, node_i
-                ):
-                    matrix[i][j] = interface_name
-                    logger.info(
-                        f"Interface name for {connection_ip} on {node_i.node_id}: {interface_name}"
-                    )
+            for conn in cycle_digraph.get_all_connections_between(node_i, node_j):
+                if isinstance(conn, RDMAConnection):
+                    matrix[i][j] = conn.source_rdma_iface
                     break
             else:
                 logger.warning(
-                    f"Failed to find interface name between {node_i.node_id} and {node_j.node_id}"
+                    f"Failed to find interface name between {node_i} and {node_j}"
                 )
                 raise ValueError(
-                    "Current ibv backend requires all-to-all rdma connections"
+                    "Current jaccl backend requires all-to-all RDMA connections"
                 )
 
     return matrix
 
 
 def _find_connection_ip(
-    node_i: NodeInfo,
-    node_j: NodeInfo,
+    node_i: NodeId,
+    node_j: NodeId,
     cycle_digraph: Topology,
 ) -> Generator[tuple[str, bool]]:
-    """Find all IP addresses that connect node i to node j, with thunderbolt flag."""
-    for connection in cycle_digraph.list_connections():
-        if (
-            connection.local_node_id == node_i.node_id
-            and connection.send_back_node_id == node_j.node_id
-        ):
-            yield connection.send_back_multiaddr.ip_address, connection.is_thunderbolt()
-
-
-def _find_rdma_interface_name_for_ip(
-    ip_address: str,
-    node_info: NodeInfo,
-) -> str | None:
-    if node_info.node_profile is None:
-        return None
-
-    logger.info(f"Searching {node_info.node_id} for ip {ip_address}:")
-    for interface in node_info.node_profile.network_interfaces:
-        if interface.name not in ["en2", "en3", "en4", "en5", "en6", "en7"]:
-            continue
-        logger.info(f" | {interface.name}: {interface.ip_address}")
-        if interface.ip_address != ip_address:
-            continue
-
-        logger.info("Found")
-        return f"rdma_{interface.name}"
-
-    return None
+    """Find all IP addresses that connect node i to node j."""
+    for connection in cycle_digraph.get_all_connections_between(node_i, node_j):
+        if isinstance(connection, SocketConnection):
+            yield connection.sink_multiaddr.ip_address, connection.is_thunderbolt()
 
 
 def _find_interface_name_for_ip(
-    ip_address: str,
-    node_info: NodeInfo,
+    ip_address: str, node_profile: NodePerformanceProfile
 ) -> str | None:
     """Find the interface name for an IP address on a node (any interface)."""
-    if node_info.node_profile is None:
-        return None
-
-    for interface in node_info.node_profile.network_interfaces:
+    for interface in node_profile.network_interfaces:
         if interface.ip_address == ip_address:
             return interface.name
 
@@ -287,7 +299,10 @@ def _find_interface_name_for_ip(
 
 
 def _find_ip_prioritised(
-    node: NodeInfo, other_node: NodeInfo, cycle_digraph: Topology
+    node_id: NodeId,
+    other_node_id: NodeId,
+    cycle_digraph: Topology,
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
 ) -> str | None:
     # TODO: Actually prioritize in the correct Ethernet > Wifi > Non-TB > TB order.
     """Find an IP address between nodes with prioritization.
@@ -298,9 +313,12 @@ def _find_ip_prioritised(
     3. Non-Thunderbolt connections
     4. Any other IP address
     """
-    ips = list(_find_connection_ip(node, other_node, cycle_digraph))
+    ips = list(_find_connection_ip(node_id, other_node_id, cycle_digraph))
     # We expect a unique iface -> ip mapping
-    iface_map = {_find_interface_name_for_ip(ip, other_node): ip for ip, _ in ips}
+    iface_map = {
+        _find_interface_name_for_ip(ip, node_profiles[other_node_id]): ip
+        for ip, _ in ips
+    }
 
     en0_ip = iface_map.get("en0")
     if en0_ip:
@@ -324,9 +342,10 @@ def _find_ip_prioritised(
 
 
 def get_mlx_ring_hosts_by_node(
-    selected_cycle: list[NodeInfo],
+    selected_cycle: Cycle,
     cycle_digraph: Topology,
     ephemeral_port: int,
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
 ) -> dict[NodeId, list[Host]]:
     """Generate per-node host lists for MLX ring backend.
 
@@ -341,14 +360,13 @@ def get_mlx_ring_hosts_by_node(
 
     hosts_by_node: dict[NodeId, list[Host]] = {}
 
-    for rank, node in enumerate(selected_cycle):
-        node_id = node.node_id
+    for rank, node_id in enumerate(selected_cycle):
         left_rank = (rank - 1) % world_size
         right_rank = (rank + 1) % world_size
 
         hosts_for_node: list[Host] = []
 
-        for idx, other_node in enumerate(selected_cycle):
+        for idx, other_node_id in enumerate(selected_cycle):
             if idx == rank:
                 hosts_for_node.append(Host(ip="0.0.0.0", port=ephemeral_port))
                 continue
@@ -358,10 +376,12 @@ def get_mlx_ring_hosts_by_node(
                 hosts_for_node.append(Host(ip="198.51.100.1", port=0))
                 continue
 
-            connection_ip = _find_ip_prioritised(node, other_node, cycle_digraph)
+            connection_ip = _find_ip_prioritised(
+                node_id, other_node_id, cycle_digraph, node_profiles
+            )
             if connection_ip is None:
                 logger.warning(
-                    f"Failed to find prioritised connection IP between {node_id} and {other_node.node_id}"
+                    f"Failed to find prioritised connection IP between {node_id} and {other_node_id}"
                 )
                 raise ValueError(
                     "MLX ring backend requires connectivity between neighbouring nodes"
@@ -375,31 +395,34 @@ def get_mlx_ring_hosts_by_node(
 
 
 def get_mlx_jaccl_coordinators(
-    selected_cycle: list[NodeInfo],
+    coordinator: NodeId,
     coordinator_port: int,
     cycle_digraph: Topology,
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
 ) -> dict[NodeId, str]:
-    """Get the coordinator addresses for MLX Jaccl (rank 0 device).
+    """Get the coordinator addresses for MLX JACCL (rank 0 device).
 
     Select an IP address that each node can reach for the rank 0 node. Returns
     address in format "X.X.X.X:PORT" per node.
     """
-    rank_0_node = selected_cycle[0]
-    logger.debug(f"Selecting coordinator from rank 0 node: {rank_0_node.node_id}")
+    logger.info(f"Selecting coordinator: {coordinator}")
 
-    def get_ip_for_node(n: NodeInfo) -> str:
-        if n.node_id == rank_0_node.node_id:
+    def get_ip_for_node(n: NodeId) -> str:
+        if n == coordinator:
             return "0.0.0.0"
 
-        ip = _find_ip_prioritised(n, rank_0_node, cycle_digraph)
-        if ip:
+        ip = _find_ip_prioritised(n, coordinator, cycle_digraph, node_profiles)
+        if ip is not None:
             return ip
 
         logger.warning(
-            f"Failed to find directly connected ip between {n.node_id} and {rank_0_node.node_id}"
+            f"Failed to find directly connected ip between {n} and {coordinator}"
+        )
+        raise ValueError(
+            "Current jaccl backend requires all participating devices to be able to communicate"
         )
-        raise ValueError("Current ibv backend requires all-to-all rdma connections")
 
     return {
-        n.node_id: f"{get_ip_for_node(n)}:{coordinator_port}" for n in selected_cycle
+        n: f"{get_ip_for_node(n)}:{coordinator_port}"
+        for n in cycle_digraph.list_nodes()
     }

src/exo/master/tests/conftest.py

@@ -1,67 +1,39 @@
-from typing import Callable
-
-import pytest
-
-from exo.shared.types.common import NodeId
 from exo.shared.types.multiaddr import Multiaddr
 from exo.shared.types.profiling import (
-    MemoryPerformanceProfile,
+    MemoryUsage,
+    NetworkInterfaceInfo,
     NodePerformanceProfile,
     SystemPerformanceProfile,
 )
-from exo.shared.types.topology import Connection, ConnectionProfile, NodeInfo
+from exo.shared.types.topology import RDMAConnection, SocketConnection
 
 
-@pytest.fixture
-def create_node():
-    def _create_node(memory: int, node_id: NodeId | None = None) -> NodeInfo:
-        if node_id is None:
-            node_id = NodeId()
-        return NodeInfo(
-            node_id=node_id,
-            node_profile=NodePerformanceProfile(
-                model_id="test",
-                chip_id="test",
-                friendly_name="test",
-                memory=MemoryPerformanceProfile.from_bytes(
-                    ram_total=1000,
-                    ram_available=memory,
-                    swap_total=1000,
-                    swap_available=1000,
-                ),
-                network_interfaces=[],
-                system=SystemPerformanceProfile(),
-            ),
-        )
-
-    return _create_node
+def create_node_profile(memory: int) -> NodePerformanceProfile:
+    return NodePerformanceProfile(
+        model_id="test",
+        chip_id="test",
+        friendly_name="test",
+        memory=MemoryUsage.from_bytes(
+            ram_total=1000,
+            ram_available=memory,
+            swap_total=1000,
+            swap_available=1000,
+        ),
+        network_interfaces=[
+            NetworkInterfaceInfo(name="en0", ip_address=f"169.254.0.{i}")
+            for i in range(10)
+        ],
+        system=SystemPerformanceProfile(),
+    )
 
 
-# TODO: this is a hack to get the port for the send_back_multiaddr
-@pytest.fixture
-def create_connection() -> Callable[[NodeId, NodeId, int | None], Connection]:
-    port_counter = 1235
-    ip_counter = 1
+def create_socket_connection(ip: int, sink_port: int = 1234) -> SocketConnection:
+    return SocketConnection(
+        sink_multiaddr=Multiaddr(address=f"/ip4/169.254.0.{ip}/tcp/{sink_port}"),
+    )
 
-    def _create_connection(
-        source_node_id: NodeId, sink_node_id: NodeId, send_back_port: int | None = None
-    ) -> Connection:
-        nonlocal port_counter
-        nonlocal ip_counter
-        # assign unique ips
-        ip_counter += 1
-        if send_back_port is None:
-            send_back_port = port_counter
-            port_counter += 1
-        return Connection(
-            local_node_id=source_node_id,
-            send_back_node_id=sink_node_id,
-            send_back_multiaddr=Multiaddr(
-                address=f"/ip4/169.254.0.{ip_counter}/tcp/{send_back_port}"
-            ),
-            connection_profile=ConnectionProfile(
-                throughput=1000, latency=1000, jitter=1000
-            ),
-        )
 
-    return _create_connection
+def create_rdma_connection(iface: int) -> RDMAConnection:
+    return RDMAConnection(
+        source_rdma_iface=f"rdma_en{iface}", sink_rdma_iface=f"rdma_en{iface}"
+    )

src/exo/master/tests/test_api_error_handling.py

@@ -0,0 +1,107 @@
+# pyright: reportUnusedFunction=false, reportAny=false
+from typing import Any, get_args
+
+from fastapi import FastAPI, HTTPException
+from fastapi.testclient import TestClient
+
+from exo.shared.types.api import ErrorInfo, ErrorResponse, FinishReason
+from exo.shared.types.chunks import TokenChunk
+from exo.worker.tests.constants import MODEL_A_ID
+
+
+def test_http_exception_handler_formats_openai_style() -> None:
+    """Test that HTTPException is converted to OpenAI-style error format."""
+    from exo.master.api import API
+
+    app = FastAPI()
+
+    # Setup exception handler
+    api = object.__new__(API)
+    api.app = app
+    api._setup_exception_handlers()  # pyright: ignore[reportPrivateUsage]
+
+    # Add test routes that raise HTTPException
+    @app.get("/test-error")
+    async def _test_error() -> None:
+        raise HTTPException(status_code=500, detail="Test error message")
+
+    @app.get("/test-not-found")
+    async def _test_not_found() -> None:
+        raise HTTPException(status_code=404, detail="Resource not found")
+
+    client = TestClient(app)
+
+    # Test 500 error
+    response = client.get("/test-error")
+    assert response.status_code == 500
+    data: dict[str, Any] = response.json()
+    assert "error" in data
+    assert data["error"]["message"] == "Test error message"
+    assert data["error"]["type"] == "Internal Server Error"
+    assert data["error"]["code"] == 500
+
+    # Test 404 error
+    response = client.get("/test-not-found")
+    assert response.status_code == 404
+    data = response.json()
+    assert "error" in data
+    assert data["error"]["message"] == "Resource not found"
+    assert data["error"]["type"] == "Not Found"
+    assert data["error"]["code"] == 404
+
+
+def test_finish_reason_includes_error() -> None:
+    valid_reasons = get_args(FinishReason)
+    assert "error" in valid_reasons
+
+
+def test_token_chunk_with_error_fields() -> None:
+    chunk = TokenChunk(
+        idx=0,
+        model=MODEL_A_ID,
+        text="",
+        token_id=0,
+        finish_reason="error",
+        error_message="Something went wrong",
+    )
+
+    assert chunk.finish_reason == "error"
+    assert chunk.error_message == "Something went wrong"
+
+
+def test_token_chunk_without_error() -> None:
+    chunk = TokenChunk(
+        idx=1,
+        model=MODEL_A_ID,
+        text="Hello",
+        token_id=42,
+        finish_reason=None,
+    )
+
+    assert chunk.finish_reason is None
+    assert chunk.error_message is None
+
+
+def test_error_response_construction() -> None:
+    error_response = ErrorResponse(
+        error=ErrorInfo(
+            message="Generation failed",
+            type="InternalServerError",
+            code=500,
+        )
+    )
+
+    assert error_response.error.message == "Generation failed"
+    assert error_response.error.code == 500
+
+
+def test_normal_finish_reasons_still_work() -> None:
+    for reason in ["stop", "length", "tool_calls", "content_filter", "function_call"]:
+        chunk = TokenChunk(
+            idx=0,
+            model=MODEL_A_ID,
+            text="done",
+            token_id=100,
+            finish_reason=reason,  # type: ignore[arg-type]
+        )
+        assert chunk.finish_reason == reason

src/exo/master/tests/test_master.py

@@ -19,15 +19,13 @@ from exo.shared.types.events import (
     ForwarderEvent,
     IndexedEvent,
     InstanceCreated,
-    NodePerformanceMeasured,
+    NodeGatheredInfo,
     TaskCreated,
 )
 from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelId, ModelMetadata
 from exo.shared.types.profiling import (
-    MemoryPerformanceProfile,
-    NodePerformanceProfile,
-    SystemPerformanceProfile,
+    MemoryUsage,
 )
 from exo.shared.types.tasks import ChatCompletion as ChatCompletionTask
 from exo.shared.types.tasks import TaskStatus
@@ -83,21 +81,14 @@ async def test_master():
                 origin=sender_node_id,
                 session=session_id,
                 event=(
-                    NodePerformanceMeasured(
+                    NodeGatheredInfo(
                         when=str(datetime.now(tz=timezone.utc)),
                         node_id=node_id,
-                        node_profile=NodePerformanceProfile(
-                            model_id="maccy",
-                            chip_id="arm",
-                            friendly_name="test",
-                            memory=MemoryPerformanceProfile(
-                                ram_total=Memory.from_bytes(678948 * 1024),
-                                ram_available=Memory.from_bytes(678948 * 1024),
-                                swap_total=Memory.from_bytes(0),
-                                swap_available=Memory.from_bytes(0),
-                            ),
-                            network_interfaces=[],
-                            system=SystemPerformanceProfile(),
+                        info=MemoryUsage(
+                            ram_total=Memory.from_bytes(678948 * 1024),
+                            ram_available=Memory.from_bytes(678948 * 1024),
+                            swap_total=Memory.from_bytes(0),
+                            swap_available=Memory.from_bytes(0),
                         ),
                     )
                 ),
@@ -163,7 +154,7 @@ async def test_master():
         assert events[0].idx == 0
         assert events[1].idx == 1
         assert events[2].idx == 2
-        assert isinstance(events[0].event, NodePerformanceMeasured)
+        assert isinstance(events[0].event, NodeGatheredInfo)
         assert isinstance(events[1].event, InstanceCreated)
         created_instance = events[1].event.instance
         assert isinstance(created_instance, MlxRingInstance)

src/exo/master/tests/test_placement.py

@@ -1,20 +1,23 @@
-from typing import Callable
-
 import pytest
-from loguru import logger
 
 from exo.master.placement import (
     get_transition_events,
     place_instance,
 )
+from exo.master.tests.conftest import (
+    create_node_profile,
+    create_rdma_connection,
+    create_socket_connection,
+)
 from exo.shared.topology import Topology
 from exo.shared.types.commands import PlaceInstance
 from exo.shared.types.common import CommandId, NodeId
 from exo.shared.types.events import InstanceCreated, InstanceDeleted
 from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelId, ModelMetadata
-from exo.shared.types.profiling import NetworkInterfaceInfo, NodePerformanceProfile
-from exo.shared.types.topology import Connection, NodeInfo
+from exo.shared.types.multiaddr import Multiaddr
+from exo.shared.types.profiling import NetworkInterfaceInfo
+from exo.shared.types.topology import Connection, SocketConnection
 from exo.shared.types.worker.instances import (
     Instance,
     InstanceId,
@@ -26,11 +29,6 @@ from exo.shared.types.worker.runners import ShardAssignments
 from exo.shared.types.worker.shards import Sharding
 
 
-@pytest.fixture
-def topology() -> Topology:
-    return Topology()
-
-
 @pytest.fixture
 def instance() -> Instance:
     return MlxRingInstance(
@@ -70,41 +68,64 @@ def place_instance_command(model_meta: ModelMetadata) -> PlaceInstance:
     [
         ((500, 500, 1000), 12, (3, 3, 6)),
         ((500, 500, 500), 12, (4, 4, 4)),
-        ((312, 518, 1024), 12, (2, 3, 7)),
+        ((312, 468, 1092), 12, (2, 3, 7)),
     ],
 )
 def test_get_instance_placements_create_instance(
     available_memory: tuple[int, int, int],
     total_layers: int,
     expected_layers: tuple[int, int, int],
-    topology: Topology,
     model_meta: ModelMetadata,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
 ):
     # arrange
     model_meta.n_layers = total_layers
     model_meta.storage_size.in_bytes = sum(
         available_memory
     )  # make it exactly fit across all nodes
+    topology = Topology()
 
     cic = place_instance_command(model_meta)
     node_id_a = NodeId()
     node_id_b = NodeId()
     node_id_c = NodeId()
-    topology.add_node(create_node(available_memory[0], node_id_a))
-    topology.add_node(create_node(available_memory[1], node_id_b))
-    topology.add_node(create_node(available_memory[2], node_id_c))
-    # Add bidirectional connections for ring topology
-    topology.add_connection(create_connection(node_id_a, node_id_b))
-    topology.add_connection(create_connection(node_id_b, node_id_a))
-    topology.add_connection(create_connection(node_id_b, node_id_c))
-    topology.add_connection(create_connection(node_id_c, node_id_b))
-    topology.add_connection(create_connection(node_id_c, node_id_a))
-    topology.add_connection(create_connection(node_id_a, node_id_c))
+
+    # fully connected (directed) between the 3 nodes
+    conn_a_b = Connection(
+        source=node_id_a, sink=node_id_b, edge=create_socket_connection(1)
+    )
+    conn_b_c = Connection(
+        source=node_id_b, sink=node_id_c, edge=create_socket_connection(2)
+    )
+    conn_c_a = Connection(
+        source=node_id_c, sink=node_id_a, edge=create_socket_connection(3)
+    )
+    conn_c_b = Connection(
+        source=node_id_c, sink=node_id_b, edge=create_socket_connection(4)
+    )
+    conn_a_c = Connection(
+        source=node_id_a, sink=node_id_c, edge=create_socket_connection(5)
+    )
+    conn_b_a = Connection(
+        source=node_id_b, sink=node_id_a, edge=create_socket_connection(6)
+    )
+
+    profiles = {
+        node_id_a: create_node_profile(available_memory[0]),
+        node_id_b: create_node_profile(available_memory[1]),
+        node_id_c: create_node_profile(available_memory[2]),
+    }
+    topology.add_node(node_id_a)
+    topology.add_node(node_id_b)
+    topology.add_node(node_id_c)
+    topology.add_connection(conn_a_b)
+    topology.add_connection(conn_b_c)
+    topology.add_connection(conn_c_a)
+    topology.add_connection(conn_c_b)
+    topology.add_connection(conn_a_c)
+    topology.add_connection(conn_b_a)
 
     # act
-    placements = place_instance(cic, topology, {})
+    placements = place_instance(cic, topology, {}, profiles)
 
     # assert
     assert len(placements) == 1
@@ -130,12 +151,11 @@ def test_get_instance_placements_create_instance(
     assert shards_sorted[-1].end_layer == total_layers
 
 
-def test_get_instance_placements_one_node_exact_fit(
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-) -> None:
+def test_get_instance_placements_one_node_exact_fit() -> None:
     topology = Topology()
     node_id = NodeId()
-    topology.add_node(create_node(1000 * 1024, node_id))
+    topology.add_node(node_id)
+    profiles = {node_id: create_node_profile(1000 * 1024)}
     cic = place_instance_command(
         ModelMetadata(
             model_id=ModelId("test-model"),
@@ -146,7 +166,7 @@ def test_get_instance_placements_one_node_exact_fit(
             supports_tensor=True,
         ),
     )
-    placements = place_instance(cic, topology, {})
+    placements = place_instance(cic, topology, {}, profiles)
 
     assert len(placements) == 1
     instance_id = list(placements.keys())[0]
@@ -157,12 +177,11 @@ def test_get_instance_placements_one_node_exact_fit(
     assert len(instance.shard_assignments.runner_to_shard) == 1
 
 
-def test_get_instance_placements_one_node_fits_with_extra_memory(
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-) -> None:
+def test_get_instance_placements_one_node_fits_with_extra_memory() -> None:
     topology = Topology()
     node_id = NodeId()
-    topology.add_node(create_node(1001 * 1024, node_id))
+    topology.add_node(node_id)
+    profiles = {node_id: create_node_profile(1001 * 1024)}
     cic = place_instance_command(
         ModelMetadata(
             model_id=ModelId("test-model"),
@@ -173,7 +192,7 @@ def test_get_instance_placements_one_node_fits_with_extra_memory(
             supports_tensor=True,
         ),
     )
-    placements = place_instance(cic, topology, {})
+    placements = place_instance(cic, topology, {}, profiles)
 
     assert len(placements) == 1
     instance_id = list(placements.keys())[0]
@@ -184,12 +203,11 @@ def test_get_instance_placements_one_node_fits_with_extra_memory(
     assert len(instance.shard_assignments.runner_to_shard) == 1
 
 
-def test_get_instance_placements_one_node_not_fit(
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-) -> None:
+def test_get_instance_placements_one_node_not_fit() -> None:
     topology = Topology()
     node_id = NodeId()
-    topology.add_node(create_node(1000 * 1024, node_id))
+    topology.add_node(node_id)
+    profiles = {node_id: create_node_profile(1000 * 1024)}
     cic = place_instance_command(
         model_meta=ModelMetadata(
             model_id=ModelId("test-model"),
@@ -202,7 +220,7 @@ def test_get_instance_placements_one_node_not_fit(
     )
 
     with pytest.raises(ValueError, match="No cycles found with sufficient memory"):
-        place_instance(cic, topology, {})
+        place_instance(cic, topology, {}, profiles)
 
 
 def test_get_transition_events_no_change(instance: Instance):
@@ -247,179 +265,130 @@ def test_get_transition_events_delete_instance(instance: Instance):
     assert events[0].instance_id == instance_id
 
 
-def test_placement_selects_cycle_with_most_memory(
-    topology: Topology,
+def test_placement_selects_leaf_nodes(
     model_meta: ModelMetadata,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
 ):
-    # Arrange two 3-node cycles with different total memory.
-    # With bidirectional connections for ring topology, both cycles have non-leaf nodes.
-    # The algorithm should select the cycle with the most available memory.
+    # arrange
+    topology = Topology()
 
-    # Model requires more than any single node but fits within a 3-node cycle
-    model_meta.storage_size.in_bytes = 1500
-    model_meta.n_layers = 12
+    model_meta.storage_size = Memory.from_bytes(1000)
 
-    # Create node ids
     node_id_a = NodeId()
     node_id_b = NodeId()
     node_id_c = NodeId()
     node_id_d = NodeId()
-    node_id_e = NodeId()
-    node_id_f = NodeId()
 
-    # A-B-C cycle total memory = 1600 (< D-E-F total)
-    topology.add_node(create_node(400, node_id_a))
-    topology.add_node(create_node(400, node_id_b))
-    topology.add_node(create_node(800, node_id_c))
+    profiles = {
+        node_id_a: create_node_profile(500),
+        node_id_b: create_node_profile(600),
+        node_id_c: create_node_profile(600),
+        node_id_d: create_node_profile(500),
+    }
 
-    # D-E-F cycle total memory = 1800 (> A-B-C total)
-    topology.add_node(create_node(600, node_id_d))
-    topology.add_node(create_node(600, node_id_e))
-    topology.add_node(create_node(600, node_id_f))
+    topology.add_node(node_id_a)
+    topology.add_node(node_id_b)
+    topology.add_node(node_id_c)
+    topology.add_node(node_id_d)
 
-    # Build bidirectional cycles for ring topology
-    topology.add_connection(create_connection(node_id_a, node_id_b))
-    topology.add_connection(create_connection(node_id_b, node_id_a))
-    topology.add_connection(create_connection(node_id_b, node_id_c))
-    topology.add_connection(create_connection(node_id_c, node_id_b))
-    topology.add_connection(create_connection(node_id_c, node_id_a))
-    topology.add_connection(create_connection(node_id_a, node_id_c))
-
-    topology.add_connection(create_connection(node_id_d, node_id_e))
-    topology.add_connection(create_connection(node_id_e, node_id_d))
-    topology.add_connection(create_connection(node_id_e, node_id_f))
-    topology.add_connection(create_connection(node_id_f, node_id_e))
-    topology.add_connection(create_connection(node_id_f, node_id_d))
-    topology.add_connection(create_connection(node_id_d, node_id_f))
-
-    cic = place_instance_command(
-        model_meta=model_meta,
+    # Daisy chain topology (directed)
+    topology.add_connection(
+        Connection(source=node_id_a, sink=node_id_b, edge=create_socket_connection(1))
+    )
+    topology.add_connection(
+        Connection(source=node_id_b, sink=node_id_a, edge=create_socket_connection(1))
+    )
+    topology.add_connection(
+        Connection(source=node_id_b, sink=node_id_c, edge=create_socket_connection(1))
+    )
+    topology.add_connection(
+        Connection(source=node_id_c, sink=node_id_b, edge=create_socket_connection(1))
+    )
+    topology.add_connection(
+        Connection(source=node_id_c, sink=node_id_d, edge=create_socket_connection(1))
+    )
+    topology.add_connection(
+        Connection(source=node_id_d, sink=node_id_c, edge=create_socket_connection(1))
     )
 
-    # Act
-    placements = place_instance(cic, topology, {})
+    cic = place_instance_command(model_meta=model_meta)
 
-    # Assert: D-E-F cycle should be selected as it has more total memory
+    # act
+    placements = place_instance(cic, topology, {}, profiles)
+
+    # assert
     assert len(placements) == 1
-    instance_id = list(placements.keys())[0]
-    instance = placements[instance_id]
+    instance = list(placements.values())[0]
 
     assigned_nodes = set(instance.shard_assignments.node_to_runner.keys())
-    less_memory_cycle_nodes = {node_id_a, node_id_b, node_id_c}
-    more_memory_cycle_nodes = {node_id_d, node_id_e, node_id_f}
-
-    assert more_memory_cycle_nodes.issubset(assigned_nodes)
-    assert assigned_nodes.isdisjoint(less_memory_cycle_nodes)
+    assert assigned_nodes == set((node_id_a, node_id_b)) or assigned_nodes == set(
+        (
+            node_id_c,
+            node_id_d,
+        )
+    )
 
 
 def test_tensor_rdma_backend_connectivity_matrix(
-    topology: Topology,
     model_meta: ModelMetadata,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
 ):
+    # arrange
+    topology = Topology()
     model_meta.n_layers = 12
     model_meta.storage_size.in_bytes = 1500
 
-    node_id_a = NodeId()
-    node_id_b = NodeId()
-    node_id_c = NodeId()
+    node_a = NodeId()
+    node_b = NodeId()
+    node_c = NodeId()
 
-    node_a = create_node(500, node_id_a)
-    node_b = create_node(500, node_id_b)
-    node_c = create_node(500, node_id_c)
+    profiles = {
+        node_a: create_node_profile(500),
+        node_b: create_node_profile(500),
+        node_c: create_node_profile(500),
+    }
 
     ethernet_interface = NetworkInterfaceInfo(
         name="en0",
-        ip_address="192.168.1.100",
+        ip_address="10.0.0.1",
+    )
+    ethernet_conn = SocketConnection(
+        sink_multiaddr=Multiaddr(address="/ip4/10.0.0.1/tcp/8000")
     )
 
-    assert node_a.node_profile is not None
-    assert node_b.node_profile is not None
-    assert node_c.node_profile is not None
-
-    conn_a_b = create_connection(node_id_a, node_id_b)
-    conn_b_c = create_connection(node_id_b, node_id_c)
-    conn_c_a = create_connection(node_id_c, node_id_a)
-
-    conn_b_a = create_connection(node_id_b, node_id_a)
-    conn_c_b = create_connection(node_id_c, node_id_b)
-    conn_a_c = create_connection(node_id_a, node_id_c)
-
-    assert conn_a_b.send_back_multiaddr is not None
-    assert conn_b_c.send_back_multiaddr is not None
-    assert conn_c_a.send_back_multiaddr is not None
-
-    assert conn_b_a.send_back_multiaddr is not None
-    assert conn_c_b.send_back_multiaddr is not None
-    assert conn_a_c.send_back_multiaddr is not None
-
-    node_a.node_profile = NodePerformanceProfile(
-        model_id="test",
-        chip_id="test",
-        friendly_name="test",
-        memory=node_a.node_profile.memory,
-        network_interfaces=[
-            NetworkInterfaceInfo(
-                name="en3",
-                ip_address=conn_c_a.send_back_multiaddr.ip_address,
-            ),
-            NetworkInterfaceInfo(
-                name="en4",
-                ip_address=conn_b_a.send_back_multiaddr.ip_address,
-            ),
-            ethernet_interface,
-        ],
-        system=node_a.node_profile.system,
-    )
-    node_b.node_profile = NodePerformanceProfile(
-        model_id="test",
-        chip_id="test",
-        friendly_name="test",
-        memory=node_b.node_profile.memory,
-        network_interfaces=[
-            NetworkInterfaceInfo(
-                name="en3",
-                ip_address=conn_c_b.send_back_multiaddr.ip_address,
-            ),
-            NetworkInterfaceInfo(
-                name="en4",
-                ip_address=conn_a_b.send_back_multiaddr.ip_address,
-            ),
-            ethernet_interface,
-        ],
-        system=node_b.node_profile.system,
-    )
-    node_c.node_profile = NodePerformanceProfile(
-        model_id="test",
-        chip_id="test",
-        friendly_name="test",
-        memory=node_c.node_profile.memory,
-        network_interfaces=[
-            NetworkInterfaceInfo(
-                name="en3",
-                ip_address=conn_a_c.send_back_multiaddr.ip_address,
-            ),
-            NetworkInterfaceInfo(
-                name="en4",
-                ip_address=conn_b_c.send_back_multiaddr.ip_address,
-            ),
-            ethernet_interface,
-        ],
-        system=node_c.node_profile.system,
-    )
+    profiles[node_a].network_interfaces = [ethernet_interface]
+    profiles[node_b].network_interfaces = [ethernet_interface]
+    profiles[node_c].network_interfaces = [ethernet_interface]
 
     topology.add_node(node_a)
     topology.add_node(node_b)
     topology.add_node(node_c)
-    topology.add_connection(conn_a_b)
-    topology.add_connection(conn_b_c)
-    topology.add_connection(conn_c_a)
-    topology.add_connection(conn_b_a)
-    topology.add_connection(conn_c_b)
-    topology.add_connection(conn_a_c)
+
+    # RDMA connections (directed)
+    topology.add_connection(
+        Connection(source=node_a, sink=node_b, edge=create_rdma_connection(3))
+    )
+    topology.add_connection(
+        Connection(source=node_b, sink=node_a, edge=create_rdma_connection(3))
+    )
+    topology.add_connection(
+        Connection(source=node_b, sink=node_c, edge=create_rdma_connection(4))
+    )
+    topology.add_connection(
+        Connection(source=node_c, sink=node_b, edge=create_rdma_connection(4))
+    )
+    topology.add_connection(
+        Connection(source=node_a, sink=node_c, edge=create_rdma_connection(5))
+    )
+    topology.add_connection(
+        Connection(source=node_c, sink=node_a, edge=create_rdma_connection(5))
+    )
+
+    # Ethernet connections (directed)
+    topology.add_connection(Connection(source=node_a, sink=node_b, edge=ethernet_conn))
+    topology.add_connection(Connection(source=node_b, sink=node_c, edge=ethernet_conn))
+    topology.add_connection(Connection(source=node_c, sink=node_a, edge=ethernet_conn))
+    topology.add_connection(Connection(source=node_a, sink=node_c, edge=ethernet_conn))
+    topology.add_connection(Connection(source=node_b, sink=node_a, edge=ethernet_conn))
+    topology.add_connection(Connection(source=node_c, sink=node_b, edge=ethernet_conn))
 
     cic = PlaceInstance(
         sharding=Sharding.Tensor,
@@ -429,35 +398,34 @@ def test_tensor_rdma_backend_connectivity_matrix(
         min_nodes=1,
     )
 
-    placements = place_instance(cic, topology, {})
+    # act
+    placements = place_instance(cic, topology, {}, profiles)
 
+    # assert
     assert len(placements) == 1
     instance_id = list(placements.keys())[0]
     instance = placements[instance_id]
 
     assert isinstance(instance, MlxJacclInstance)
 
-    assert instance.ibv_devices is not None
+    assert instance.jaccl_devices is not None
     assert instance.jaccl_coordinators is not None
 
-    matrix = instance.ibv_devices
+    matrix = instance.jaccl_devices
     assert len(matrix) == 3
-
     for i in range(3):
         assert matrix[i][i] is None
 
     assigned_nodes = list(instance.shard_assignments.node_to_runner.keys())
     node_to_idx = {node_id: idx for idx, node_id in enumerate(assigned_nodes)}
 
-    idx_a = node_to_idx[node_id_a]
-    idx_b = node_to_idx[node_id_b]
-    idx_c = node_to_idx[node_id_c]
+    idx_a = node_to_idx[node_a]
+    idx_b = node_to_idx[node_b]
+    idx_c = node_to_idx[node_c]
 
-    logger.info(matrix)
-
-    assert matrix[idx_a][idx_b] == "rdma_en4"
-    assert matrix[idx_b][idx_c] == "rdma_en3"
-    assert matrix[idx_c][idx_a] == "rdma_en3"
+    assert matrix[idx_a][idx_b] == "rdma_en3"
+    assert matrix[idx_b][idx_c] == "rdma_en4"
+    assert matrix[idx_c][idx_a] == "rdma_en5"
 
     # Verify coordinators are set for all nodes
     assert len(instance.jaccl_coordinators) == 3
@@ -469,7 +437,5 @@ def test_tensor_rdma_backend_connectivity_matrix(
         if node_id == assigned_nodes[0]:
             assert coordinator.startswith("0.0.0.0:")
         else:
-            # Non-rank-0 nodes should have valid IP addresses (can be link-local)
             ip_part = coordinator.split(":")[0]
-            # Just verify it's a valid IP format
             assert len(ip_part.split(".")) == 4

src/exo/master/tests/test_placement_utils.py

@@ -1,162 +1,187 @@
-from typing import Callable
+from copy import copy
 
 import pytest
 
 from exo.master.placement_utils import (
+    allocate_layers_proportionally,
     filter_cycles_by_memory,
     get_hosts_from_subgraph,
     get_mlx_jaccl_coordinators,
     get_shard_assignments,
     get_smallest_cycles,
 )
+from exo.master.tests.conftest import create_node_profile, create_socket_connection
 from exo.shared.topology import Topology
 from exo.shared.types.common import Host, NodeId
 from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelId, ModelMetadata
-from exo.shared.types.profiling import NetworkInterfaceInfo, NodePerformanceProfile
-from exo.shared.types.topology import Connection, NodeInfo
+from exo.shared.types.profiling import (
+    MemoryUsage,
+    NetworkInterfaceInfo,
+    NodePerformanceProfile,
+    SystemPerformanceProfile,
+)
+from exo.shared.types.topology import Connection, SocketConnection
 from exo.shared.types.worker.shards import Sharding
 
 
-@pytest.fixture
-def topology() -> Topology:
-    topology = Topology()
-    return topology
-
-
-def test_filter_cycles_by_memory(
-    topology: Topology,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
-):
+def test_filter_cycles_by_memory():
     # arrange
     node1_id = NodeId()
     node2_id = NodeId()
+    connection1 = Connection(
+        source=node1_id, sink=node2_id, edge=create_socket_connection(1)
+    )
+    connection2 = Connection(
+        source=node2_id, sink=node1_id, edge=create_socket_connection(2)
+    )
 
-    node1 = create_node(1000 * 1024, node1_id)
-    node2 = create_node(1000 * 1024, node2_id)
-
-    topology.add_node(node1)
-    topology.add_node(node2)
-
-    connection1 = create_connection(node1_id, node2_id)
-    connection2 = create_connection(node2_id, node1_id)
+    node1 = create_node_profile(1000 * 1024)
+    node2 = create_node_profile(1000 * 1024)
+    node_profiles = {node1_id: node1, node2_id: node2}
 
+    topology = Topology()
+    topology.add_node(node1_id)
+    topology.add_node(node2_id)
     topology.add_connection(connection1)
     topology.add_connection(connection2)
 
-    cycles = topology.get_cycles()
+    cycles = [c for c in topology.get_cycles() if len(c) != 1]
     assert len(cycles) == 1
     assert len(cycles[0]) == 2
 
     # act
-    filtered_cycles = filter_cycles_by_memory(cycles, Memory.from_bytes(1))
+    filtered_cycles = filter_cycles_by_memory(
+        cycles, node_profiles, Memory.from_bytes(1)
+    )
 
     # assert
     assert len(filtered_cycles) == 1
     assert len(filtered_cycles[0]) == 2
-    assert set(n.node_id for n in filtered_cycles[0]) == {node1_id, node2_id}
+    assert set(n for n in filtered_cycles[0]) == {node1_id, node2_id}
 
 
-def test_filter_cycles_by_insufficient_memory(
-    topology: Topology,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
-):
+def test_filter_cycles_by_insufficient_memory():
     # arrange
     node1_id = NodeId()
     node2_id = NodeId()
+    connection1 = Connection(
+        source=node1_id, sink=node2_id, edge=create_socket_connection(1)
+    )
+    connection2 = Connection(
+        source=node2_id, sink=node1_id, edge=create_socket_connection(2)
+    )
 
-    node1 = create_node(1000 * 1024, node1_id)
-    node2 = create_node(1000 * 1024, node2_id)
-
-    topology.add_node(node1)
-    topology.add_node(node2)
-
-    connection1 = create_connection(node1_id, node2_id)
-    connection2 = create_connection(node2_id, node1_id)
+    node1 = create_node_profile(1000 * 1024)
+    node2 = create_node_profile(1000 * 1024)
+    node_profiles = {node1_id: node1, node2_id: node2}
 
+    topology = Topology()
+    topology.add_node(node1_id)
+    topology.add_node(node2_id)
     topology.add_connection(connection1)
     topology.add_connection(connection2)
 
     # act
     filtered_cycles = filter_cycles_by_memory(
-        topology.get_cycles(), Memory.from_kb(2001)
+        topology.get_cycles(), node_profiles, Memory.from_kb(2001)
     )
 
     # assert
     assert len(filtered_cycles) == 0
 
 
-def test_filter_multiple_cycles_by_memory(
-    topology: Topology,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
-):
+def test_filter_multiple_cycles_by_memory():
     # arrange
     node_a_id = NodeId()
     node_b_id = NodeId()
     node_c_id = NodeId()
+    connection1 = Connection(
+        source=node_a_id, sink=node_b_id, edge=create_socket_connection(1)
+    )
+    connection2 = Connection(
+        source=node_b_id, sink=node_a_id, edge=create_socket_connection(2)
+    )
+    connection3 = Connection(
+        source=node_a_id, sink=node_c_id, edge=create_socket_connection(3)
+    )
+    connection4 = Connection(
+        source=node_c_id, sink=node_b_id, edge=create_socket_connection(4)
+    )
 
-    node_a = create_node(500 * 1024, node_a_id)
-    node_b = create_node(500 * 1024, node_b_id)
-    node_c = create_node(1000 * 1024, node_c_id)
+    node_a = create_node_profile(500 * 1024)
+    node_b = create_node_profile(500 * 1024)
+    node_c = create_node_profile(1000 * 1024)
+    node_profiles = {
+        node_a_id: node_a,
+        node_b_id: node_b,
+        node_c_id: node_c,
+    }
 
-    topology.add_node(node_a)
-    topology.add_node(node_b)
-    topology.add_node(node_c)
-
-    topology.add_connection(create_connection(node_a_id, node_b_id))
-    topology.add_connection(create_connection(node_b_id, node_a_id))
-
-    topology.add_connection(create_connection(node_a_id, node_c_id))
-    topology.add_connection(create_connection(node_c_id, node_b_id))
+    topology = Topology()
+    topology.add_node(node_a_id)
+    topology.add_node(node_b_id)
+    topology.add_node(node_c_id)
+    topology.add_connection(connection1)
+    topology.add_connection(connection2)
+    topology.add_connection(connection3)
+    topology.add_connection(connection4)
 
     cycles = topology.get_cycles()
 
     # act
-    filtered_cycles = filter_cycles_by_memory(cycles, Memory.from_kb(1500))
+    filtered_cycles = filter_cycles_by_memory(
+        cycles, node_profiles, Memory.from_kb(1500)
+    )
 
     # assert
     assert len(filtered_cycles) == 1
     assert len(filtered_cycles[0]) == 3
-    assert set(n.node_id for n in filtered_cycles[0]) == {
+    assert set(n for n in filtered_cycles[0]) == {
         node_a_id,
         node_b_id,
         node_c_id,
     }
 
 
-def test_get_smallest_cycles(
-    topology: Topology,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
-):
+def test_get_smallest_cycles():
     # arrange
     node_a_id = NodeId()
     node_b_id = NodeId()
     node_c_id = NodeId()
 
-    node_a = create_node(500 * 1024, node_a_id)
-    node_b = create_node(500 * 1024, node_b_id)
-    node_c = create_node(1000 * 1024, node_c_id)
+    topology = Topology()
+    topology.add_node(node_a_id)
+    topology.add_node(node_b_id)
+    topology.add_node(node_c_id)
 
-    topology.add_node(node_a)
-    topology.add_node(node_b)
-    topology.add_node(node_c)
+    connection1 = Connection(
+        source=node_a_id, sink=node_b_id, edge=create_socket_connection(1)
+    )
+    connection2 = Connection(
+        source=node_b_id, sink=node_a_id, edge=create_socket_connection(2)
+    )
+    connection3 = Connection(
+        source=node_a_id, sink=node_c_id, edge=create_socket_connection(3)
+    )
+    connection4 = Connection(
+        source=node_c_id, sink=node_b_id, edge=create_socket_connection(4)
+    )
 
-    topology.add_connection(create_connection(node_a_id, node_b_id))
-    topology.add_connection(create_connection(node_b_id, node_c_id))
-    topology.add_connection(create_connection(node_c_id, node_a_id))
-    topology.add_connection(create_connection(node_b_id, node_a_id))
+    topology.add_connection(connection1)
+    topology.add_connection(connection2)
+    topology.add_connection(connection3)
+    topology.add_connection(connection4)
+
+    cycles = [c for c in topology.get_cycles() if len(c) != 1]  # ignore singletons
 
     # act
-    smallest_cycles = get_smallest_cycles(topology.get_cycles())
+    smallest_cycles = get_smallest_cycles(cycles)
 
     # assert
     assert len(smallest_cycles) == 1
     assert len(smallest_cycles[0]) == 2
-    assert set(n.node_id for n in smallest_cycles[0]) == {node_a_id, node_b_id}
+    assert set(n for n in smallest_cycles[0]) == {node_a_id, node_b_id}
 
 
 @pytest.mark.parametrize(
@@ -165,12 +190,12 @@ def test_get_smallest_cycles(
         ((500, 500, 1000), 12, (3, 3, 6)),
         ((500, 500, 500), 12, (4, 4, 4)),
         ((312, 518, 1024), 12, (2, 3, 7)),
+        # Edge case: one node has ~90% of memory - should not over-allocate.
+        # Each node must have enough memory for at least 1 layer (50 KB = 1000/20).
+        ((900, 50, 50), 20, (18, 1, 1)),
     ],
 )
 def test_get_shard_assignments(
-    topology: Topology,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId], Connection],
     available_memory: tuple[int, int, int],
     total_layers: int,
     expected_layers: tuple[int, int, int],
@@ -180,18 +205,37 @@ def test_get_shard_assignments(
     node_b_id = NodeId()
     node_c_id = NodeId()
 
-    node_a = create_node(available_memory[0] * 1024, node_a_id)
-    node_b = create_node(available_memory[1] * 1024, node_b_id)
-    node_c = create_node(available_memory[2] * 1024, node_c_id)
+    # create connections (A -> B -> C -> A forms a 3-cycle, plus B -> A also exists)
+    connection1 = Connection(
+        source=node_a_id, sink=node_b_id, edge=create_socket_connection(1)
+    )
+    connection2 = Connection(
+        source=node_b_id, sink=node_c_id, edge=create_socket_connection(2)
+    )
+    connection3 = Connection(
+        source=node_c_id, sink=node_a_id, edge=create_socket_connection(3)
+    )
+    connection4 = Connection(
+        source=node_b_id, sink=node_a_id, edge=create_socket_connection(4)
+    )
 
-    topology.add_node(node_a)
-    topology.add_node(node_b)
-    topology.add_node(node_c)
+    topology = Topology()
+    topology.add_node(node_a_id)
+    topology.add_node(node_b_id)
+    topology.add_node(node_c_id)
+    topology.add_connection(connection1)
+    topology.add_connection(connection2)
+    topology.add_connection(connection3)
+    topology.add_connection(connection4)
 
-    topology.add_connection(create_connection(node_a_id, node_b_id))
-    topology.add_connection(create_connection(node_b_id, node_c_id))
-    topology.add_connection(create_connection(node_c_id, node_a_id))
-    topology.add_connection(create_connection(node_b_id, node_a_id))
+    node_a = create_node_profile(available_memory[0] * 1024)
+    node_b = create_node_profile(available_memory[1] * 1024)
+    node_c = create_node_profile(available_memory[2] * 1024)
+    node_profiles = {
+        node_a_id: node_a,
+        node_b_id: node_b,
+        node_c_id: node_c,
+    }
 
     model_meta = ModelMetadata(
         model_id=ModelId("test-model"),
@@ -201,23 +245,22 @@ def test_get_shard_assignments(
         hidden_size=1000,
         supports_tensor=True,
     )
+
     cycles = topology.get_cycles()
-    selected_cycle = cycles[0]
+
+    # pick the 3-node cycle deterministically (cycle ordering can vary)
+    selected_cycle = next(cycle for cycle in cycles if len(cycle) == 3)
 
     # act
     shard_assignments = get_shard_assignments(
-        model_meta, selected_cycle, Sharding.Pipeline
+        model_meta, selected_cycle, Sharding.Pipeline, node_profiles=node_profiles
     )
 
     # assert
     runner_id_a = shard_assignments.node_to_runner[node_a_id]
     runner_id_b = shard_assignments.node_to_runner[node_b_id]
     runner_id_c = shard_assignments.node_to_runner[node_c_id]
-    assert (
-        shard_assignments.runner_to_shard[runner_id_c].end_layer
-        - shard_assignments.runner_to_shard[runner_id_c].start_layer
-        == expected_layers[2]
-    )
+
     assert (
         shard_assignments.runner_to_shard[runner_id_a].end_layer
         - shard_assignments.runner_to_shard[runner_id_a].start_layer
@@ -228,30 +271,37 @@ def test_get_shard_assignments(
         - shard_assignments.runner_to_shard[runner_id_b].start_layer
         == expected_layers[1]
     )
+    assert (
+        shard_assignments.runner_to_shard[runner_id_c].end_layer
+        - shard_assignments.runner_to_shard[runner_id_c].start_layer
+        == expected_layers[2]
+    )
 
 
-def test_get_hosts_from_subgraph(
-    topology: Topology,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId, int | None], Connection],
-):
+def test_get_hosts_from_subgraph():
     # arrange
     node_a_id = NodeId()
     node_b_id = NodeId()
     node_c_id = NodeId()
+    topology = Topology()
 
-    node_a = create_node(500, node_a_id)
-    node_b = create_node(500, node_b_id)
-    node_c = create_node(1000, node_c_id)
+    topology.add_node(node_a_id)
+    topology.add_node(node_b_id)
+    topology.add_node(node_c_id)
 
-    topology.add_node(node_a)
-    topology.add_node(node_b)
-    topology.add_node(node_c)
+    connection1 = Connection(
+        source=node_a_id, sink=node_b_id, edge=create_socket_connection(1)
+    )
+    connection2 = Connection(
+        source=node_b_id, sink=node_c_id, edge=create_socket_connection(2)
+    )
+    connection3 = Connection(
+        source=node_c_id, sink=node_a_id, edge=create_socket_connection(3)
+    )
 
-    topology.add_connection(create_connection(node_a_id, node_b_id, 5001))
-    topology.add_connection(create_connection(node_b_id, node_c_id, 5002))
-    topology.add_connection(create_connection(node_c_id, node_a_id, 5003))
-    topology.add_connection(create_connection(node_b_id, node_a_id, 5004))
+    topology.add_connection(connection1)
+    topology.add_connection(connection2)
+    topology.add_connection(connection3)
 
     # act
     hosts = get_hosts_from_subgraph(topology)
@@ -259,95 +309,78 @@ def test_get_hosts_from_subgraph(
     # assert
     assert len(hosts) == 3
     expected_hosts = [
-        Host(ip=("169.254.0.2"), port=5001),
-        Host(ip=("169.254.0.3"), port=5002),
-        Host(ip=("169.254.0.4"), port=5003),
+        Host(ip="169.254.0.1", port=1234),
+        Host(ip="169.254.0.2", port=1234),
+        Host(ip="169.254.0.3", port=1234),
     ]
     for expected_host in expected_hosts:
         assert expected_host in hosts
 
 
-def test_get_mlx_jaccl_coordinators(
-    topology: Topology,
-    create_node: Callable[[int, NodeId | None], NodeInfo],
-    create_connection: Callable[[NodeId, NodeId, int | None], Connection],
-):
+def test_get_mlx_jaccl_coordinators():
     # arrange
     node_a_id = NodeId()
     node_b_id = NodeId()
     node_c_id = NodeId()
 
-    node_a = create_node(500 * 1024, node_a_id)
-    node_b = create_node(500 * 1024, node_b_id)
-    node_c = create_node(1000 * 1024, node_c_id)
+    # fully connected (directed) between the 3 nodes
+    conn_a_b = Connection(
+        source=node_a_id, sink=node_b_id, edge=create_socket_connection(1)
+    )
+    conn_b_a = Connection(
+        source=node_b_id, sink=node_a_id, edge=create_socket_connection(2)
+    )
+    conn_b_c = Connection(
+        source=node_b_id, sink=node_c_id, edge=create_socket_connection(3)
+    )
+    conn_c_b = Connection(
+        source=node_c_id, sink=node_b_id, edge=create_socket_connection(4)
+    )
+    conn_c_a = Connection(
+        source=node_c_id, sink=node_a_id, edge=create_socket_connection(5)
+    )
+    conn_a_c = Connection(
+        source=node_a_id, sink=node_c_id, edge=create_socket_connection(6)
+    )
 
-    conn_a_b = create_connection(node_a_id, node_b_id, 5001)
-    conn_b_a = create_connection(node_b_id, node_a_id, 5002)
-    conn_b_c = create_connection(node_b_id, node_c_id, 5003)
-    conn_c_b = create_connection(node_c_id, node_b_id, 5004)
-    conn_c_a = create_connection(node_c_id, node_a_id, 5005)
-    conn_a_c = create_connection(node_a_id, node_c_id, 5006)
-
-    # Update node profiles with network interfaces before adding to topology
-    assert node_a.node_profile is not None
-    assert node_b.node_profile is not None
-    assert node_c.node_profile is not None
-
-    node_a.node_profile = NodePerformanceProfile(
+    npp = NodePerformanceProfile(
         model_id="test",
         chip_id="test",
         friendly_name="test",
-        memory=node_a.node_profile.memory,
-        network_interfaces=[
-            NetworkInterfaceInfo(
-                name="en3",
-                ip_address=conn_a_b.send_back_multiaddr.ip_address,
-            ),
-            NetworkInterfaceInfo(
-                name="en4",
-                ip_address=conn_a_c.send_back_multiaddr.ip_address,
-            ),
-        ],
-        system=node_a.node_profile.system,
-    )
-    node_b.node_profile = NodePerformanceProfile(
-        model_id="test",
-        chip_id="test",
-        friendly_name="test",
-        memory=node_b.node_profile.memory,
-        network_interfaces=[
-            NetworkInterfaceInfo(
-                name="en3",
-                ip_address=conn_b_a.send_back_multiaddr.ip_address,
-            ),
-            NetworkInterfaceInfo(
-                name="en4",
-                ip_address=conn_b_c.send_back_multiaddr.ip_address,
-            ),
-        ],
-        system=node_b.node_profile.system,
-    )
-    node_c.node_profile = NodePerformanceProfile(
-        model_id="test",
-        chip_id="test",
-        friendly_name="test",
-        memory=node_c.node_profile.memory,
-        network_interfaces=[
-            NetworkInterfaceInfo(
-                name="en3",
-                ip_address=conn_c_b.send_back_multiaddr.ip_address,
-            ),
-            NetworkInterfaceInfo(
-                name="en4",
-                ip_address=conn_c_a.send_back_multiaddr.ip_address,
-            ),
-        ],
-        system=node_c.node_profile.system,
+        memory=MemoryUsage.from_bytes(
+            ram_total=0,
+            ram_available=0,
+            swap_total=0,
+            swap_available=0,
+        ),
+        network_interfaces=[],
+        system=SystemPerformanceProfile(),
     )
+    npp_a = copy(npp)
+    npp_a.network_interfaces = [
+        NetworkInterfaceInfo(name="en0", ip_address="169.254.0.5"),
+        NetworkInterfaceInfo(name="en0", ip_address="169.254.0.2"),
+    ]
+    npp_b = copy(npp)
+    npp_b.network_interfaces = [
+        NetworkInterfaceInfo(name="en0", ip_address="169.254.0.1"),
+        NetworkInterfaceInfo(name="en0", ip_address="169.254.0.4"),
+    ]
+    npp_c = copy(npp)
+    npp_c.network_interfaces = [
+        NetworkInterfaceInfo(name="en0", ip_address="169.254.0.3"),
+        NetworkInterfaceInfo(name="en0", ip_address="169.254.0.6"),
+    ]
+    node_profiles = {
+        node_a_id: npp_a,
+        node_b_id: npp_b,
+        node_c_id: npp_c,
+    }
 
-    topology.add_node(node_a)
-    topology.add_node(node_b)
-    topology.add_node(node_c)
+    topology = Topology()
+    topology.add_node(node_a_id)
+    topology.add_node(node_b_id)
+    topology.add_node(node_c_id)
 
     topology.add_connection(conn_a_b)
     topology.add_connection(conn_b_a)
@@ -356,11 +389,12 @@ def test_get_mlx_jaccl_coordinators(
     topology.add_connection(conn_c_a)
     topology.add_connection(conn_a_c)
 
-    cycle = [node_a, node_b, node_c]
-
     # act
     coordinators = get_mlx_jaccl_coordinators(
-        cycle, coordinator_port=5000, cycle_digraph=topology
+        node_a_id,
+        coordinator_port=5000,
+        cycle_digraph=topology,
+        node_profiles=node_profiles,
     )
 
     # assert
@@ -381,19 +415,128 @@ def test_get_mlx_jaccl_coordinators(
             f"Coordinator for {node_id} should use port 5000"
         )
 
-    # Rank 0 (node_a) treats this as the listen socket so should listen on all
-    # IPs
+    # Rank 0 (node_a) treats this as the listen socket so should listen on all IPs
     assert coordinators[node_a_id].startswith("0.0.0.0:"), (
-        "Rank 0 node should use localhost as coordinator"
+        "Rank 0 node should use 0.0.0.0 as coordinator listen address"
     )
 
     # Non-rank-0 nodes should use the specific IP from their connection to rank 0
     # node_b uses the IP from conn_b_a (node_b -> node_a)
-    assert coordinators[node_b_id] == (
-        f"{conn_b_a.send_back_multiaddr.ip_address}:5000"
+    assert isinstance(conn_b_a.edge, SocketConnection)
+    assert (
+        coordinators[node_b_id] == f"{conn_b_a.edge.sink_multiaddr.ip_address}:5000"
     ), "node_b should use the IP from conn_b_a"
 
     # node_c uses the IP from conn_c_a (node_c -> node_a)
+    assert isinstance(conn_c_a.edge, SocketConnection)
     assert coordinators[node_c_id] == (
-        f"{conn_c_a.send_back_multiaddr.ip_address}:5000"
+        f"{conn_c_a.edge.sink_multiaddr.ip_address}:5000"
     ), "node_c should use the IP from conn_c_a"
+
+
+class TestAllocateLayersProportionally:
+    def test_empty_node_list_raises(self):
+        with pytest.raises(ValueError, match="empty node list"):
+            allocate_layers_proportionally(total_layers=10, memory_fractions=[])
+
+    def test_zero_layers_raises(self):
+        with pytest.raises(ValueError, match="need at least 1 layer per node"):
+            allocate_layers_proportionally(total_layers=0, memory_fractions=[0.5, 0.5])
+
+    def test_negative_layers_raises(self):
+        with pytest.raises(ValueError, match="need at least 1 layer per node"):
+            allocate_layers_proportionally(total_layers=-1, memory_fractions=[0.5, 0.5])
+
+    def test_fewer_layers_than_nodes_raises(self):
+        with pytest.raises(ValueError, match="need at least 1 layer per node"):
+            allocate_layers_proportionally(
+                total_layers=2, memory_fractions=[0.33, 0.33, 0.34]
+            )
+
+    def test_equal_distribution(self):
+        result = allocate_layers_proportionally(
+            total_layers=12, memory_fractions=[0.25, 0.25, 0.25, 0.25]
+        )
+        assert result == [3, 3, 3, 3]
+        assert sum(result) == 12
+
+    def test_proportional_distribution(self):
+        result = allocate_layers_proportionally(
+            total_layers=12, memory_fractions=[0.25, 0.25, 0.50]
+        )
+        assert result == [3, 3, 6]
+        assert sum(result) == 12
+
+    def test_extreme_imbalance_ensures_minimum(self):
+        result = allocate_layers_proportionally(
+            total_layers=20, memory_fractions=[0.975, 0.0125, 0.0125]
+        )
+        assert all(layers >= 1 for layers in result)
+        assert sum(result) == 20
+        # Small nodes get minimum 1 layer
+        assert result == [18, 1, 1]
+
+    def test_single_node_gets_all_layers(self):
+        result = allocate_layers_proportionally(total_layers=10, memory_fractions=[1.0])
+        assert result == [10]
+
+    def test_minimum_viable_allocation(self):
+        result = allocate_layers_proportionally(
+            total_layers=3, memory_fractions=[0.33, 0.33, 0.34]
+        )
+        assert result == [1, 1, 1]
+        assert sum(result) == 3
+
+
+def test_get_shard_assignments_insufficient_memory_raises():
+    """Test that ValueError is raised when a node has insufficient memory for its layers."""
+    node_a_id = NodeId()
+    node_b_id = NodeId()
+    node_c_id = NodeId()
+    topology = Topology()
+
+    # Node C has only 10 KB but would need 50 KB for 1 layer (1000 KB / 20 layers)
+    node_a = create_node_profile(900 * 1024)
+    node_b = create_node_profile(50 * 1024)
+    node_c = create_node_profile(10 * 1024)  # Insufficient memory
+
+    topology.add_node(node_a_id)
+    topology.add_node(node_b_id)
+    topology.add_node(node_c_id)
+
+    conn_a_b = Connection(
+        source=node_a_id, sink=node_b_id, edge=create_socket_connection(1)
+    )
+    conn_b_c = Connection(
+        source=node_b_id, sink=node_c_id, edge=create_socket_connection(2)
+    )
+    conn_c_a = Connection(
+        source=node_c_id, sink=node_a_id, edge=create_socket_connection(3)
+    )
+    conn_b_a = Connection(
+        source=node_b_id, sink=node_a_id, edge=create_socket_connection(3)
+    )
+    topology.add_connection(conn_a_b)
+    topology.add_connection(conn_b_c)
+    topology.add_connection(conn_c_a)
+    topology.add_connection(conn_b_a)
+
+    profiles = {
+        node_a_id: node_a,
+        node_b_id: node_b,
+        node_c_id: node_c,
+    }
+
+    model_meta = ModelMetadata(
+        model_id=ModelId("test-model"),
+        pretty_name="Test Model",
+        n_layers=20,
+        storage_size=Memory.from_kb(1000),
+        hidden_size=1000,
+        supports_tensor=True,
+    )
+    cycles = topology.get_cycles()
+    selected_cycle = cycles[0]
+
+    with pytest.raises(ValueError, match="insufficient memory"):
+        get_shard_assignments(model_meta, selected_cycle, Sharding.Pipeline, profiles)

src/exo/master/tests/test_topology.py

@@ -1,13 +1,14 @@
 import pytest
 
 from exo.shared.topology import Topology
+from exo.shared.types.common import NodeId
 from exo.shared.types.multiaddr import Multiaddr
 from exo.shared.types.profiling import (
-    MemoryPerformanceProfile,
+    MemoryUsage,
     NodePerformanceProfile,
     SystemPerformanceProfile,
 )
-from exo.shared.types.topology import Connection, ConnectionProfile, NodeId, NodeInfo
+from exo.shared.types.topology import Connection, SocketConnection
 
 
 @pytest.fixture
@@ -16,20 +17,15 @@ def topology() -> Topology:
 
 
 @pytest.fixture
-def connection() -> Connection:
-    return Connection(
-        local_node_id=NodeId(),
-        send_back_node_id=NodeId(),
-        send_back_multiaddr=Multiaddr(address="/ip4/127.0.0.1/tcp/1235"),
-        connection_profile=ConnectionProfile(
-            throughput=1000, latency=1000, jitter=1000
-        ),
+def socket_connection() -> SocketConnection:
+    return SocketConnection(
+        sink_multiaddr=Multiaddr(address="/ip4/127.0.0.1/tcp/1235"),
     )
 
 
 @pytest.fixture
 def node_profile() -> NodePerformanceProfile:
-    memory_profile = MemoryPerformanceProfile.from_bytes(
+    memory_profile = MemoryUsage.from_bytes(
         ram_total=1000, ram_available=1000, swap_total=1000, swap_available=1000
     )
     system_profile = SystemPerformanceProfile()
@@ -43,162 +39,91 @@ def node_profile() -> NodePerformanceProfile:
     )
 
 
-@pytest.fixture
-def connection_profile() -> ConnectionProfile:
-    return ConnectionProfile(throughput=1000, latency=1000, jitter=1000)
-
-
-def test_add_node(topology: Topology, node_profile: NodePerformanceProfile):
+def test_add_node(topology: Topology):
     # arrange
     node_id = NodeId()
 
     # act
-    topology.add_node(NodeInfo(node_id=node_id, node_profile=node_profile))
+    topology.add_node(node_id)
 
     # assert
-    data = topology.get_node_profile(node_id)
-    assert data == node_profile
+    assert topology.node_is_leaf(node_id)
 
 
-def test_add_connection(
-    topology: Topology, node_profile: NodePerformanceProfile, connection: Connection
-):
+def test_add_connection(topology: Topology, socket_connection: SocketConnection):
     # arrange
-    topology.add_node(
-        NodeInfo(node_id=connection.local_node_id, node_profile=node_profile)
-    )
-    topology.add_node(
-        NodeInfo(node_id=connection.send_back_node_id, node_profile=node_profile)
-    )
+    node_a = NodeId()
+    node_b = NodeId()
+    connection = Connection(source=node_a, sink=node_b, edge=socket_connection)
+
+    topology.add_node(node_a)
+    topology.add_node(node_b)
     topology.add_connection(connection)
 
     # act
-    data = topology.get_connection_profile(connection)
+    data = list(topology.list_connections())
 
     # assert
-    assert data == connection.connection_profile
+    assert data == [connection]
 
-
-def test_update_node_profile(
-    topology: Topology, node_profile: NodePerformanceProfile, connection: Connection
-):
-    # arrange
-    topology.add_node(
-        NodeInfo(node_id=connection.local_node_id, node_profile=node_profile)
-    )
-    topology.add_node(
-        NodeInfo(node_id=connection.send_back_node_id, node_profile=node_profile)
-    )
-    topology.add_connection(connection)
-
-    new_node_profile = NodePerformanceProfile(
-        model_id="test",
-        chip_id="test",
-        friendly_name="test",
-        memory=MemoryPerformanceProfile.from_bytes(
-            ram_total=1000, ram_available=1000, swap_total=1000, swap_available=1000
-        ),
-        network_interfaces=[],
-        system=SystemPerformanceProfile(),
-    )
-
-    # act
-    topology.update_node_profile(
-        connection.local_node_id, node_profile=new_node_profile
-    )
-
-    # assert
-    data = topology.get_node_profile(connection.local_node_id)
-    assert data == new_node_profile
-
-
-def test_update_connection_profile(
-    topology: Topology, node_profile: NodePerformanceProfile, connection: Connection
-):
-    # arrange
-    topology.add_node(
-        NodeInfo(node_id=connection.local_node_id, node_profile=node_profile)
-    )
-    topology.add_node(
-        NodeInfo(node_id=connection.send_back_node_id, node_profile=node_profile)
-    )
-    topology.add_connection(connection)
-
-    new_connection_profile = ConnectionProfile(
-        throughput=2000, latency=2000, jitter=2000
-    )
-    connection = Connection(
-        local_node_id=connection.local_node_id,
-        send_back_node_id=connection.send_back_node_id,
-        send_back_multiaddr=connection.send_back_multiaddr,
-        connection_profile=new_connection_profile,
-    )
-
-    # act
-    topology.update_connection_profile(connection)
-
-    # assert
-    data = topology.get_connection_profile(connection)
-    assert data == new_connection_profile
+    assert topology.node_is_leaf(node_a)
+    assert topology.node_is_leaf(node_b)
 
 
 def test_remove_connection_still_connected(
-    topology: Topology, node_profile: NodePerformanceProfile, connection: Connection
+    topology: Topology, socket_connection: SocketConnection
 ):
     # arrange
-    topology.add_node(
-        NodeInfo(node_id=connection.local_node_id, node_profile=node_profile)
-    )
-    topology.add_node(
-        NodeInfo(node_id=connection.send_back_node_id, node_profile=node_profile)
-    )
-    topology.add_connection(connection)
+    node_a = NodeId()
+    node_b = NodeId()
+    conn = Connection(source=node_a, sink=node_b, edge=socket_connection)
+
+    topology.add_node(node_a)
+    topology.add_node(node_b)
+    topology.add_connection(conn)
 
     # act
-    topology.remove_connection(connection)
+    topology.remove_connection(conn)
 
     # assert
-    assert topology.get_connection_profile(connection) is None
+    assert list(topology.get_all_connections_between(node_a, node_b)) == []
 
 
 def test_remove_node_still_connected(
-    topology: Topology, node_profile: NodePerformanceProfile, connection: Connection
+    topology: Topology, socket_connection: SocketConnection
 ):
     # arrange
-    topology.add_node(
-        NodeInfo(node_id=connection.local_node_id, node_profile=node_profile)
-    )
-    topology.add_node(
-        NodeInfo(node_id=connection.send_back_node_id, node_profile=node_profile)
-    )
-    topology.add_connection(connection)
+    node_a = NodeId()
+    node_b = NodeId()
+    conn = Connection(source=node_a, sink=node_b, edge=socket_connection)
+
+    topology.add_node(node_a)
+    topology.add_node(node_b)
+    topology.add_connection(conn)
+    assert list(topology.out_edges(node_a)) == [conn]
 
     # act
-    topology.remove_node(connection.local_node_id)
+    topology.remove_node(node_b)
 
     # assert
-    assert topology.get_node_profile(connection.local_node_id) is None
+    assert list(topology.out_edges(node_a)) == []
 
 
-def test_list_nodes(
-    topology: Topology, node_profile: NodePerformanceProfile, connection: Connection
-):
+def test_list_nodes(topology: Topology, socket_connection: SocketConnection):
     # arrange
-    topology.add_node(
-        NodeInfo(node_id=connection.local_node_id, node_profile=node_profile)
-    )
-    topology.add_node(
-        NodeInfo(node_id=connection.send_back_node_id, node_profile=node_profile)
-    )
-    topology.add_connection(connection)
+    node_a = NodeId()
+    node_b = NodeId()
+    conn = Connection(source=node_a, sink=node_b, edge=socket_connection)
+
+    topology.add_node(node_a)
+    topology.add_node(node_b)
+    topology.add_connection(conn)
+    assert list(topology.out_edges(node_a)) == [conn]
 
     # act
     nodes = list(topology.list_nodes())
 
     # assert
     assert len(nodes) == 2
-    assert all(isinstance(node, NodeInfo) for node in nodes)
-    assert {node.node_id for node in nodes} == {
-        connection.local_node_id,
-        connection.send_back_node_id,
-    }
+    assert all(isinstance(node, NodeId) for node in nodes)
+    assert set(node for node in nodes) == set([node_a, node_b])

src/exo/shared/apply.py

@@ -11,10 +11,8 @@ from exo.shared.types.events import (
     IndexedEvent,
     InstanceCreated,
     InstanceDeleted,
-    NodeCreated,
     NodeDownloadProgress,
-    NodeMemoryMeasured,
-    NodePerformanceMeasured,
+    NodeGatheredInfo,
     NodeTimedOut,
     RunnerDeleted,
     RunnerStatusUpdated,
@@ -27,13 +25,23 @@ from exo.shared.types.events import (
     TopologyEdgeCreated,
     TopologyEdgeDeleted,
 )
-from exo.shared.types.profiling import NodePerformanceProfile, SystemPerformanceProfile
+from exo.shared.types.profiling import NodePerformanceProfile
 from exo.shared.types.state import State
 from exo.shared.types.tasks import Task, TaskId, TaskStatus
-from exo.shared.types.topology import NodeInfo
+from exo.shared.types.topology import Connection, RDMAConnection
 from exo.shared.types.worker.downloads import DownloadProgress
 from exo.shared.types.worker.instances import Instance, InstanceId
 from exo.shared.types.worker.runners import RunnerId, RunnerStatus
+from exo.utils.info_gatherer.info_gatherer import (
+    MacmonMetrics,
+    MacThunderboltConnections,
+    MacThunderboltIdentifiers,
+    MemoryUsage,
+    MiscData,
+    NodeConfig,
+    NodeNetworkInterfaces,
+    StaticNodeInformation,
+)
 
 
 def event_apply(event: Event, state: State) -> State:
@@ -47,16 +55,12 @@ def event_apply(event: Event, state: State) -> State:
             return apply_instance_created(event, state)
         case InstanceDeleted():
             return apply_instance_deleted(event, state)
-        case NodeCreated():
-            return apply_topology_node_created(event, state)
         case NodeTimedOut():
             return apply_node_timed_out(event, state)
-        case NodePerformanceMeasured():
-            return apply_node_performance_measured(event, state)
         case NodeDownloadProgress():
             return apply_node_download_progress(event, state)
-        case NodeMemoryMeasured():
-            return apply_node_memory_measured(event, state)
+        case NodeGatheredInfo():
+            return apply_node_gathered_info(event, state)
         case RunnerDeleted():
             return apply_runner_deleted(event, state)
         case RunnerStatusUpdated():
@@ -188,7 +192,7 @@ def apply_runner_deleted(event: RunnerDeleted, state: State) -> State:
 
 
 def apply_node_timed_out(event: NodeTimedOut, state: State) -> State:
-    topology = copy.copy(state.topology)
+    topology = copy.deepcopy(state.topology)
     state.topology.remove_node(event.node_id)
     node_profiles = {
         key: value for key, value in state.node_profiles.items() if key != event.node_id
@@ -196,8 +200,12 @@ def apply_node_timed_out(event: NodeTimedOut, state: State) -> State:
     last_seen = {
         key: value for key, value in state.last_seen.items() if key != event.node_id
     }
+    downloads = {
+        key: value for key, value in state.downloads.items() if key != event.node_id
+    }
     return state.model_copy(
         update={
+            "downloads": downloads,
             "topology": topology,
             "node_profiles": node_profiles,
             "last_seen": last_seen,
@@ -205,103 +213,68 @@ def apply_node_timed_out(event: NodeTimedOut, state: State) -> State:
     )
 
 
-def apply_node_performance_measured(
-    event: NodePerformanceMeasured, state: State
-) -> State:
-    new_profiles: Mapping[NodeId, NodePerformanceProfile] = {
-        **state.node_profiles,
-        event.node_id: event.node_profile,
-    }
-    last_seen: Mapping[NodeId, datetime] = {
-        **state.last_seen,
-        event.node_id: datetime.fromisoformat(event.when),
-    }
-    state = state.model_copy(update={"node_profiles": new_profiles})
-    topology = copy.copy(state.topology)
-    # TODO: NodeCreated
-    if not topology.contains_node(event.node_id):
-        topology.add_node(NodeInfo(node_id=event.node_id))
-    topology.update_node_profile(event.node_id, event.node_profile)
+def apply_node_gathered_info(event: NodeGatheredInfo, state: State) -> State:
+    topology = copy.deepcopy(state.topology)
+    topology.add_node(event.node_id)
+    info = event.info
+    profile = state.node_profiles.get(event.node_id, NodePerformanceProfile())
+    match info:
+        case MacmonMetrics():
+            profile.system = info.system_profile
+            profile.memory = info.memory
+        case MemoryUsage():
+            profile.memory = info
+        case NodeConfig():
+            pass
+        case MiscData():
+            profile.friendly_name = info.friendly_name
+        case StaticNodeInformation():
+            profile.model_id = info.model
+            profile.chip_id = info.chip
+        case NodeNetworkInterfaces():
+            profile.network_interfaces = info.ifaces
+        case MacThunderboltIdentifiers():
+            profile.tb_interfaces = info.idents
+        case MacThunderboltConnections():
+            conn_map = {
+                tb_ident.domain_uuid: (nid, tb_ident.rdma_interface)
+                for nid in state.node_profiles
+                for tb_ident in state.node_profiles[nid].tb_interfaces
+            }
+            as_rdma_conns = [
+                Connection(
+                    source=event.node_id,
+                    sink=conn_map[tb_conn.sink_uuid][0],
+                    edge=RDMAConnection(
+                        source_rdma_iface=conn_map[tb_conn.source_uuid][1],
+                        sink_rdma_iface=conn_map[tb_conn.sink_uuid][1],
+                    ),
+                )
+                for tb_conn in info.conns
+                if tb_conn.source_uuid in conn_map
+                if tb_conn.sink_uuid in conn_map
+            ]
+            topology.replace_all_out_rdma_connections(event.node_id, as_rdma_conns)
+
+    last_seen = {**state.last_seen, event.node_id: datetime.fromisoformat(event.when)}
+    new_profiles = {**state.node_profiles, event.node_id: profile}
     return state.model_copy(
         update={
             "node_profiles": new_profiles,
-            "topology": topology,
             "last_seen": last_seen,
+            "topology": topology,
         }
     )
 
 
-def apply_node_memory_measured(event: NodeMemoryMeasured, state: State) -> State:
-    existing = state.node_profiles.get(event.node_id)
-    topology = copy.copy(state.topology)
-
-    if existing is None:
-        created = NodePerformanceProfile(
-            model_id="unknown",
-            chip_id="unknown",
-            friendly_name="Unknown",
-            memory=event.memory,
-            network_interfaces=[],
-            system=SystemPerformanceProfile(
-                # TODO: flops_fp16=0.0,
-                gpu_usage=0.0,
-                temp=0.0,
-                sys_power=0.0,
-                pcpu_usage=0.0,
-                ecpu_usage=0.0,
-                ane_power=0.0,
-            ),
-        )
-        created_profiles: Mapping[NodeId, NodePerformanceProfile] = {
-            **state.node_profiles,
-            event.node_id: created,
-        }
-        last_seen: Mapping[NodeId, datetime] = {
-            **state.last_seen,
-            event.node_id: datetime.fromisoformat(event.when),
-        }
-        if not topology.contains_node(event.node_id):
-            topology.add_node(NodeInfo(node_id=event.node_id))
-            # TODO: NodeCreated
-        topology.update_node_profile(event.node_id, created)
-        return state.model_copy(
-            update={
-                "node_profiles": created_profiles,
-                "topology": topology,
-                "last_seen": last_seen,
-            }
-        )
-
-    updated = existing.model_copy(update={"memory": event.memory})
-    updated_profiles: Mapping[NodeId, NodePerformanceProfile] = {
-        **state.node_profiles,
-        event.node_id: updated,
-    }
-    # TODO: NodeCreated
-    if not topology.contains_node(event.node_id):
-        topology.add_node(NodeInfo(node_id=event.node_id))
-    topology.update_node_profile(event.node_id, updated)
-    return state.model_copy(
-        update={"node_profiles": updated_profiles, "topology": topology}
-    )
-
-
-def apply_topology_node_created(event: NodeCreated, state: State) -> State:
-    topology = copy.copy(state.topology)
-    topology.add_node(NodeInfo(node_id=event.node_id))
-    return state.model_copy(update={"topology": topology})
-
-
 def apply_topology_edge_created(event: TopologyEdgeCreated, state: State) -> State:
-    topology = copy.copy(state.topology)
-    topology.add_connection(event.edge)
+    topology = copy.deepcopy(state.topology)
+    topology.add_connection(event.conn)
     return state.model_copy(update={"topology": topology})
 
 
 def apply_topology_edge_deleted(event: TopologyEdgeDeleted, state: State) -> State:
-    topology = copy.copy(state.topology)
-    if not topology.contains_connection(event.edge):
-        return state
-    topology.remove_connection(event.edge)
+    topology = copy.deepcopy(state.topology)
+    topology.remove_connection(event.conn)
     # TODO: Clean up removing the reverse connection
     return state.model_copy(update={"topology": topology})

src/exo/shared/constants.py

@@ -38,6 +38,7 @@ EXO_TEST_LOG = EXO_CACHE_HOME / "exo_test.log"
 
 # Identity (config)
 EXO_NODE_ID_KEYPAIR = EXO_CONFIG_HOME / "node_id.keypair"
+EXO_CONFIG_FILE = EXO_CONFIG_HOME / "config.toml"
 
 # libp2p topics for event forwarding
 LIBP2P_LOCAL_EVENTS_TOPIC = "worker_events"

src/exo/shared/logging.py

@@ -11,9 +11,6 @@ class InterceptLogger(HypercornLogger):
     def __init__(self, config: Config):
         super().__init__(config)
         assert self.error_logger
-        # TODO: Decide if we want to provide access logs
-        # assert self.access_logger
-        # self.access_logger.handlers = [_InterceptHandler()]
         self.error_logger.handlers = [_InterceptHandler()]
 
 
@@ -29,6 +26,11 @@ class _InterceptHandler(logging.Handler):
 
 def logger_setup(log_file: Path | None, verbosity: int = 0):
     """Set up logging for this process - formatting, file handles, verbosity and output"""
+
+    logging.getLogger("exo_pyo3_bindings").setLevel(logging.WARNING)
+    logging.getLogger("httpx").setLevel(logging.WARNING)
+    logging.getLogger("httpcore").setLevel(logging.WARNING)
+
     logger.remove()
 
     # replace all stdlib loggers with _InterceptHandlers that log to loguru

src/exo/shared/models/model_cards.py

@@ -14,32 +14,6 @@ class ModelCard(CamelCaseModel):
 
 MODEL_CARDS: dict[str, ModelCard] = {
     # deepseek v3
-    # "deepseek-v3-0324:4bit": ModelCard(
-    #     short_id="deepseek-v3-0324:4bit",
-    #     model_id="mlx-community/DeepSeek-V3-0324-4bit",
-    #     name="DeepSeek V3 0324 (4-bit)",
-    #     description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-V3-0324-4bit"),
-    #         pretty_name="DeepSeek V3 0324 (4-bit)",
-    #         storage_size=Memory.from_kb(409706307),
-    #         n_layers=61,
-    #     ),
-    # ),
-    # "deepseek-v3-0324": ModelCard(
-    #     short_id="deepseek-v3-0324",
-    #     model_id="mlx-community/DeepSeek-v3-0324-8bit",
-    #     name="DeepSeek V3 0324 (8-bit)",
-    #     description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-v3-0324-8bit"),
-    #         pretty_name="DeepSeek V3 0324 (8-bit)",
-    #         storage_size=Memory.from_kb(754706307),
-    #         n_layers=61,
-    #     ),
-    # ),
     "deepseek-v3.1-4bit": ModelCard(
         short_id="deepseek-v3.1-4bit",
         model_id=ModelId("mlx-community/DeepSeek-V3.1-4bit"),
@@ -70,63 +44,6 @@ MODEL_CARDS: dict[str, ModelCard] = {
             supports_tensor=True,
         ),
     ),
-    # "deepseek-v3.2": ModelCard(
-    #     short_id="deepseek-v3.2",
-    #     model_id=ModelId("mlx-community/DeepSeek-V3.2-8bit"),
-    #     name="DeepSeek V3.2 (8-bit)",
-    #     description="""DeepSeek V3.2 is a large language model trained on the DeepSeek V3.2 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-V3.2-8bit"),
-    #         pretty_name="DeepSeek V3.2 (8-bit)",
-    #         storage_size=Memory.from_kb(754706307),
-    #         n_layers=61,
-    #         hidden_size=7168,
-    #     ),
-    # ),
-    # "deepseek-v3.2-4bit": ModelCard(
-    #     short_id="deepseek-v3.2-4bit",
-    #     model_id=ModelId("mlx-community/DeepSeek-V3.2-4bit"),
-    #     name="DeepSeek V3.2 (4-bit)",
-    #     description="""DeepSeek V3.2 is a large language model trained on the DeepSeek V3.2 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-V3.2-4bit"),
-    #         pretty_name="DeepSeek V3.2 (4-bit)",
-    #         storage_size=Memory.from_kb(754706307 // 2),  # TODO !!!!!
-    #         n_layers=61,
-    #         hidden_size=7168,
-    #     ),
-    # ),
-    # deepseek r1
-    # "deepseek-r1-0528-4bit": ModelCard(
-    #     short_id="deepseek-r1-0528-4bit",
-    #     model_id="mlx-community/DeepSeek-R1-0528-4bit",
-    #     name="DeepSeek-R1-0528 (4-bit)",
-    #     description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-R1-0528-4bit"),
-    #         pretty_name="DeepSeek R1 671B (4-bit)",
-    #         storage_size=Memory.from_kb(409706307),
-    #         n_layers=61,
-    #         hidden_size=7168,
-    #     ),
-    # ),
-    # "deepseek-r1-0528": ModelCard(
-    #     short_id="deepseek-r1-0528",
-    #     model_id="mlx-community/DeepSeek-R1-0528-8bit",
-    #     name="DeepSeek-R1-0528 (8-bit)",
-    #     description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/DeepSeek-R1-0528-8bit"),
-    #         pretty_name="DeepSeek R1 671B (8-bit)",
-    #         storage_size=Memory.from_bytes(754998771712),
-    #         n_layers=61,
-    # .       hidden_size=7168,
-    #     ),
-    # ),
     # kimi k2
     "kimi-k2-instruct-4bit": ModelCard(
         short_id="kimi-k2-instruct-4bit",
@@ -508,23 +425,24 @@ MODEL_CARDS: dict[str, ModelCard] = {
             supports_tensor=True,
         ),
     ),
-    "gpt-oss-20b-4bit": ModelCard(
-        short_id="gpt-oss-20b-4bit",
-        model_id=ModelId("mlx-community/gpt-oss-20b-MXFP4-Q4"),
-        name="GPT-OSS 20B (MXFP4-Q4, MLX)",
-        description="""OpenAI's GPT-OSS 20B is a medium-sized MoE model for lower-latency and local or specialized use cases; this MLX variant uses MXFP4 4-bit quantization.""",
+    "gpt-oss-20b-MXFP4-Q8": ModelCard(
+        short_id="gpt-oss-20b-MXFP4-Q8",
+        model_id=ModelId("mlx-community/gpt-oss-20b-MXFP4-Q8"),
+        name="GPT-OSS 20B (MXFP4-Q8, MLX)",
+        description="""OpenAI's GPT-OSS 20B is a medium-sized MoE model for lower-latency and local or specialized use cases; this variant is a 4-bit MLX conversion for Apple Silicon.""",
         tags=[],
         metadata=ModelMetadata(
-            model_id=ModelId("mlx-community/gpt-oss-20b-MXFP4-Q4"),
-            pretty_name="GPT-OSS 20B (MXFP4-Q4, MLX)",
+            model_id=ModelId("mlx-community/gpt-oss-20b-MXFP4-Q8"),
+            pretty_name="GPT-OSS 20B (MXFP4-Q8, MLX)",
             storage_size=Memory.from_kb(11_744_051),
             n_layers=24,
             hidden_size=2880,
             supports_tensor=True,
         ),
     ),
-    # Needs to be quantized g32 or g16.
+    # glm 4.5
     "glm-4.5-air-8bit": ModelCard(
+        # Needs to be quantized g32 or g16 to work with tensor parallel
         short_id="glm-4.5-air-8bit",
         model_id=ModelId("mlx-community/GLM-4.5-Air-8bit"),
         name="GLM 4.5 Air 8bit",
@@ -554,19 +472,81 @@ MODEL_CARDS: dict[str, ModelCard] = {
             supports_tensor=True,
         ),
     ),
-    # "devstral-2-123b-instruct-2512-8bit": ModelCard(
-    #     short_id="devstral-2-123b-instruct-2512-8bit",
-    #     model_id=ModelId("mlx-community/Devstral-2-123B-Instruct-2512-8bit"),
-    #     name="Devstral 2 123B Instruct 2512 (8-bit, MLX)",
-    #     description="""Mistral AI's Devstral 2 123B Instruct (2512) is an agentic coding model.""",
-    #     tags=[],
-    #     metadata=ModelMetadata(
-    #         model_id=ModelId("mlx-community/Devstral-2-123B-Instruct-2512-8bit"),
-    #         pretty_name="Devstral 2 123B Instruct 2512 (8-bit, MLX)",
-    #         storage_size=Memory.from_kb(133_000_000),
-    #         n_layers=88,
-    #         hidden_size=12288,
-    #         supports_tensor=True,
-    #     ),
-    # ),
+    # glm 4.7
+    "glm-4.7-4bit": ModelCard(
+        short_id="glm-4.7-4bit",
+        model_id=ModelId("mlx-community/GLM-4.7-4bit"),
+        name="GLM 4.7 4bit",
+        description="GLM 4.7 4bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/GLM-4.7-4bit"),
+            pretty_name="GLM 4.7 4bit",
+            storage_size=Memory.from_bytes(198556925568),
+            n_layers=91,
+            hidden_size=5120,
+            supports_tensor=True,
+        ),
+    ),
+    "glm-4.7-6bit": ModelCard(
+        short_id="glm-4.7-6bit",
+        model_id=ModelId("mlx-community/GLM-4.7-6bit"),
+        name="GLM 4.7 6bit",
+        description="GLM 4.7 6bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/GLM-4.7-6bit"),
+            pretty_name="GLM 4.7 6bit",
+            storage_size=Memory.from_bytes(286737579648),
+            n_layers=91,
+            hidden_size=5120,
+            supports_tensor=True,
+        ),
+    ),
+    "glm-4.7-8bit-gs32": ModelCard(
+        short_id="glm-4.7-8bit-gs32",
+        model_id=ModelId("mlx-community/GLM-4.7-8bit-gs32"),
+        name="GLM 4.7 8bit (gs32)",
+        description="GLM 4.7 8bit (gs32)",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/GLM-4.7-8bit-gs32"),
+            pretty_name="GLM 4.7 8bit (gs32)",
+            storage_size=Memory.from_bytes(396963397248),
+            n_layers=91,
+            hidden_size=5120,
+            supports_tensor=True,
+        ),
+    ),
+    # minimax-m2
+    "minimax-m2.1-8bit": ModelCard(
+        short_id="minimax-m2.1-8bit",
+        model_id=ModelId("mlx-community/MiniMax-M2.1-8bit"),
+        name="MiniMax M2.1 8bit",
+        description="MiniMax M2.1 8bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/MiniMax-M2.1-8bit"),
+            pretty_name="MiniMax M2.1 8bit",
+            storage_size=Memory.from_bytes(242986745856),
+            n_layers=61,
+            hidden_size=3072,
+            supports_tensor=True,
+        ),
+    ),
+    "minimax-m2.1-3bit": ModelCard(
+        short_id="minimax-m2.1-3bit",
+        model_id=ModelId("mlx-community/MiniMax-M2.1-3bit"),
+        name="MiniMax M2.1 3bit",
+        description="MiniMax M2.1 3bit",
+        tags=[],
+        metadata=ModelMetadata(
+            model_id=ModelId("mlx-community/MiniMax-M2.1-3bit"),
+            pretty_name="MiniMax M2.1 3bit",
+            storage_size=Memory.from_bytes(100086644736),
+            n_layers=61,
+            hidden_size=3072,
+            supports_tensor=True,
+        ),
+    ),
 }

src/exo/shared/tests/test_apply/test_apply_node_download.py

@@ -2,6 +2,7 @@ from exo.shared.apply import apply_node_download_progress
 from exo.shared.tests.conftest import get_pipeline_shard_metadata
 from exo.shared.types.common import NodeId
 from exo.shared.types.events import NodeDownloadProgress
+from exo.shared.types.memory import Memory
 from exo.shared.types.state import State
 from exo.shared.types.worker.downloads import DownloadCompleted
 from exo.worker.tests.constants import MODEL_A_ID, MODEL_B_ID
@@ -13,6 +14,7 @@ def test_apply_node_download_progress():
     event = DownloadCompleted(
         node_id=NodeId("node-1"),
         shard_metadata=shard1,
+        total_bytes=Memory(),
     )
 
     new_state = apply_node_download_progress(
@@ -28,10 +30,12 @@ def test_apply_two_node_download_progress():
     event1 = DownloadCompleted(
         node_id=NodeId("node-1"),
         shard_metadata=shard1,
+        total_bytes=Memory(),
     )
     event2 = DownloadCompleted(
         node_id=NodeId("node-1"),
         shard_metadata=shard2,
+        total_bytes=Memory(),
     )
     state = State(downloads={NodeId("node-1"): [event1]})
 
@@ -39,7 +43,4 @@ def test_apply_two_node_download_progress():
         NodeDownloadProgress(download_progress=event2), state
     )
 
-    # TODO: This test is failing. We should support the following:
-    # 1. Downloading multiple models concurrently on the same node (one per runner is fine).
-    # 2. Downloading a model, it completes, then downloading a different model on the same node.
     assert new_state.downloads == {NodeId("node-1"): [event1, event2]}

src/exo/shared/tests/test_state_serialization.py

@@ -1,7 +1,7 @@
 from exo.shared.types.common import NodeId
 from exo.shared.types.multiaddr import Multiaddr
 from exo.shared.types.state import State
-from exo.shared.types.topology import Connection
+from exo.shared.types.topology import Connection, SocketConnection
 
 
 def test_state_serialization_roundtrip() -> None:
@@ -12,9 +12,11 @@ def test_state_serialization_roundtrip() -> None:
     node_b = NodeId("node-b")
 
     connection = Connection(
-        local_node_id=node_a,
-        send_back_node_id=node_b,
-        send_back_multiaddr=Multiaddr(address="/ip4/127.0.0.1/tcp/10001"),
+        source=node_a,
+        sink=node_b,
+        edge=SocketConnection(
+            sink_multiaddr=Multiaddr(address="/ip4/127.0.0.1/tcp/10001"),
+        ),
     )
 
     state = State()
@@ -23,5 +25,11 @@ def test_state_serialization_roundtrip() -> None:
     json_repr = state.model_dump_json()
     restored_state = State.model_validate_json(json_repr)
 
-    assert state.topology.to_snapshot() == restored_state.topology.to_snapshot()
+    assert (
+        state.topology.to_snapshot().nodes
+        == restored_state.topology.to_snapshot().nodes
+    )
+    assert set(state.topology.to_snapshot().connections) == set(
+        restored_state.topology.to_snapshot().connections
+    )
     assert restored_state.model_dump_json() == json_repr

src/exo/shared/topology.py

@@ -1,203 +1,227 @@
 import contextlib
+from collections.abc import Mapping, Sequence
+from dataclasses import dataclass, field
 from typing import Iterable
 
 import rustworkx as rx
 from pydantic import BaseModel, ConfigDict
 
 from exo.shared.types.common import NodeId
-from exo.shared.types.profiling import ConnectionProfile, NodePerformanceProfile
-from exo.shared.types.topology import Connection, NodeInfo
+from exo.shared.types.topology import (
+    Connection,
+    Cycle,
+    RDMAConnection,
+    SocketConnection,
+)
 
 
 class TopologySnapshot(BaseModel):
-    nodes: list[NodeInfo]
-    connections: list[Connection]
+    nodes: Sequence[NodeId]
+    connections: Mapping[
+        NodeId, Mapping[NodeId, Sequence[SocketConnection | RDMAConnection]]
+    ]
 
-    model_config = ConfigDict(frozen=True, extra="forbid", strict=True)
+    model_config = ConfigDict(frozen=True, extra="forbid")
 
 
+@dataclass
 class Topology:
-    def __init__(self) -> None:
-        self._graph: rx.PyDiGraph[NodeInfo, Connection] = rx.PyDiGraph()
-        self._node_id_to_rx_id_map: dict[NodeId, int] = dict()
-        self._rx_id_to_node_id_map: dict[int, NodeId] = dict()
-        self._edge_id_to_rx_id_map: dict[Connection, int] = dict()
+    _graph: rx.PyDiGraph[NodeId, SocketConnection | RDMAConnection] = field(
+        init=False, default_factory=rx.PyDiGraph
+    )
+    _vertex_indices: dict[NodeId, int] = field(init=False, default_factory=dict)
 
     def to_snapshot(self) -> TopologySnapshot:
         return TopologySnapshot(
-            nodes=list(self.list_nodes()),
-            connections=list(self.list_connections()),
+            nodes=list(self.list_nodes()), connections=self.map_connections()
         )
 
     @classmethod
     def from_snapshot(cls, snapshot: TopologySnapshot) -> "Topology":
         topology = cls()
 
-        for node in snapshot.nodes:
+        for node_id in snapshot.nodes:
             with contextlib.suppress(ValueError):
-                topology.add_node(node)
+                topology.add_node(node_id)
 
-        for connection in snapshot.connections:
-            topology.add_connection(connection)
+        for source in snapshot.connections:
+            for sink in snapshot.connections[source]:
+                for edge in snapshot.connections[source][sink]:
+                    topology.add_connection(
+                        Connection(source=source, sink=sink, edge=edge)
+                    )
 
         return topology
 
-    def add_node(self, node: NodeInfo) -> None:
-        if node.node_id in self._node_id_to_rx_id_map:
+    def add_node(self, node_id: NodeId) -> None:
+        if node_id in self._vertex_indices:
             return
-        rx_id = self._graph.add_node(node)
-        self._node_id_to_rx_id_map[node.node_id] = rx_id
-        self._rx_id_to_node_id_map[rx_id] = node.node_id
+        rx_id = self._graph.add_node(node_id)
+        self._vertex_indices[node_id] = rx_id
 
     def node_is_leaf(self, node_id: NodeId) -> bool:
         return (
-            node_id in self._node_id_to_rx_id_map
-            and len(self._graph.neighbors(self._node_id_to_rx_id_map[node_id])) == 1
+            node_id in self._vertex_indices
+            and len(self._graph.neighbors(self._vertex_indices[node_id])) <= 1
         )
 
     def neighbours(self, node_id: NodeId) -> list[NodeId]:
         return [
-            self._rx_id_to_node_id_map[rx_id]
-            for rx_id in self._graph.neighbors(self._node_id_to_rx_id_map[node_id])
+            self._graph[rx_id]
+            for rx_id in self._graph.neighbors(self._vertex_indices[node_id])
         ]
 
-    def out_edges(self, node_id: NodeId) -> list[tuple[NodeId, Connection]]:
-        if node_id not in self._node_id_to_rx_id_map:
+    def out_edges(self, node_id: NodeId) -> Iterable[Connection]:
+        if node_id not in self._vertex_indices:
             return []
-        return [
-            (self._rx_id_to_node_id_map[nid], conn)
-            for _, nid, conn in self._graph.out_edges(
-                self._node_id_to_rx_id_map[node_id]
+        return (
+            Connection(source=self._graph[source], sink=self._graph[sink], edge=edge)
+            for source, sink, edge in self._graph.out_edges(
+                self._vertex_indices[node_id]
             )
-        ]
+        )
 
     def contains_node(self, node_id: NodeId) -> bool:
-        return node_id in self._node_id_to_rx_id_map
+        return node_id in self._vertex_indices
 
-    def contains_connection(self, connection: Connection) -> bool:
-        return connection in self._edge_id_to_rx_id_map
-
-    def add_connection(
-        self,
-        connection: Connection,
-    ) -> None:
-        if connection.local_node_id not in self._node_id_to_rx_id_map:
-            self.add_node(NodeInfo(node_id=connection.local_node_id))
-        if connection.send_back_node_id not in self._node_id_to_rx_id_map:
-            self.add_node(NodeInfo(node_id=connection.send_back_node_id))
-
-        if connection in self._edge_id_to_rx_id_map:
+    def add_connection(self, conn: Connection) -> None:
+        source, sink, edge = conn.source, conn.sink, conn.edge
+        del conn
+        if edge in self.get_all_connections_between(source, sink):
             return
 
-        src_id = self._node_id_to_rx_id_map[connection.local_node_id]
-        sink_id = self._node_id_to_rx_id_map[connection.send_back_node_id]
+        if source not in self._vertex_indices:
+            self.add_node(source)
+        if sink not in self._vertex_indices:
+            self.add_node(sink)
 
-        rx_id = self._graph.add_edge(src_id, sink_id, connection)
-        self._edge_id_to_rx_id_map[connection] = rx_id
+        src_id = self._vertex_indices[source]
+        sink_id = self._vertex_indices[sink]
 
-    def list_nodes(self) -> Iterable[NodeInfo]:
-        return (self._graph[i] for i in self._graph.node_indices())
+        _ = self._graph.add_edge(src_id, sink_id, edge)
 
-    def list_connections(self) -> Iterable[Connection]:
-        return (connection for _, _, connection in self._graph.weighted_edge_list())
+    def get_all_connections_between(
+        self, source: NodeId, sink: NodeId
+    ) -> Iterable[SocketConnection | RDMAConnection]:
+        if source not in self._vertex_indices:
+            return []
+        if sink not in self._vertex_indices:
+            return []
 
-    def get_node_profile(self, node_id: NodeId) -> NodePerformanceProfile | None:
+        src_id = self._vertex_indices[source]
+        sink_id = self._vertex_indices[sink]
         try:
-            rx_idx = self._node_id_to_rx_id_map[node_id]
-            return self._graph.get_node_data(rx_idx).node_profile
-        except KeyError:
-            return None
+            return self._graph.get_all_edge_data(src_id, sink_id)
+        except rx.NoEdgeBetweenNodes:
+            return []
 
-    def update_node_profile(
-        self, node_id: NodeId, node_profile: NodePerformanceProfile
-    ) -> None:
-        rx_idx = self._node_id_to_rx_id_map[node_id]
-        self._graph[rx_idx].node_profile = node_profile
+    def list_nodes(self) -> Iterable[NodeId]:
+        return self._graph.nodes()
 
-    def update_connection_profile(self, connection: Connection) -> None:
-        rx_idx = self._edge_id_to_rx_id_map[connection]
-        self._graph.update_edge_by_index(rx_idx, connection)
+    def map_connections(
+        self,
+    ) -> Mapping[NodeId, Mapping[NodeId, Sequence[SocketConnection | RDMAConnection]]]:
+        base: dict[NodeId, dict[NodeId, list[SocketConnection | RDMAConnection]]] = {}
+        for src_id, sink_id, connection in self._graph.weighted_edge_list():
+            source = self._graph[src_id]
+            sink = self._graph[sink_id]
+            if source not in base:
+                base[source] = {}
+            if sink not in base[source]:
+                base[source][sink] = []
+            base[source][sink].append(connection)
+        return base
 
-    def get_connection_profile(
-        self, connection: Connection
-    ) -> ConnectionProfile | None:
-        try:
-            rx_idx = self._edge_id_to_rx_id_map[connection]
-            return self._graph.get_edge_data_by_index(rx_idx).connection_profile
-        except KeyError:
-            return None
+    def list_connections(
+        self,
+    ) -> Iterable[Connection]:
+        return (
+            (
+                Connection(
+                    source=self._graph[src_id],
+                    sink=self._graph[sink_id],
+                    edge=connection,
+                )
+            )
+            for src_id, sink_id, connection in self._graph.weighted_edge_list()
+        )
 
     def remove_node(self, node_id: NodeId) -> None:
-        if node_id not in self._node_id_to_rx_id_map:
+        if node_id not in self._vertex_indices:
             return
 
-        for connection in self.list_connections():
-            if (
-                connection.local_node_id == node_id
-                or connection.send_back_node_id == node_id
-            ):
-                self.remove_connection(connection)
-
-        rx_idx = self._node_id_to_rx_id_map[node_id]
+        rx_idx = self._vertex_indices[node_id]
         self._graph.remove_node(rx_idx)
 
-        del self._node_id_to_rx_id_map[node_id]
-        del self._rx_id_to_node_id_map[rx_idx]
+        del self._vertex_indices[node_id]
 
-    def remove_connection(self, connection: Connection) -> None:
-        if connection not in self._edge_id_to_rx_id_map:
+    def replace_all_out_rdma_connections(
+        self, source: NodeId, new_connections: Sequence[Connection]
+    ) -> None:
+        for conn_idx in self._graph.out_edge_indices(self._vertex_indices[source]):
+            if isinstance(self._graph.get_edge_data_by_index(conn_idx), RDMAConnection):
+                self._graph.remove_edge_from_index(conn_idx)
+        for conn in new_connections:
+            self.add_connection(conn)
+
+    def remove_connection(self, conn: Connection) -> None:
+        if (
+            conn.source not in self._vertex_indices
+            or conn.sink not in self._vertex_indices
+        ):
             return
-        rx_idx = self._edge_id_to_rx_id_map[connection]
-        self._graph.remove_edge_from_index(rx_idx)
-        del self._edge_id_to_rx_id_map[connection]
+        for conn_idx in self._graph.edge_indices_from_endpoints(
+            self._vertex_indices[conn.source], self._vertex_indices[conn.sink]
+        ):
+            if self._graph.get_edge_data_by_index(conn_idx) == conn.edge:
+                self._graph.remove_edge_from_index(conn_idx)
+
+    def get_cycles(self) -> list[Cycle]:
+        """Get simple cycles in the graph, including singleton cycles"""
 
-    def get_cycles(self) -> list[list[NodeInfo]]:
         cycle_idxs = rx.simple_cycles(self._graph)
-        cycles: list[list[NodeInfo]] = []
+        cycles: list[Cycle] = []
         for cycle_idx in cycle_idxs:
-            cycle = [self._graph[idx] for idx in cycle_idx]
+            cycle = Cycle(node_ids=[self._graph[idx] for idx in cycle_idx])
             cycles.append(cycle)
-
+        for node_id in self.list_nodes():
+            cycles.append(Cycle(node_ids=[node_id]))
         return cycles
 
-    def get_cycles_tb(self) -> list[list[NodeInfo]]:
+    def get_cycles_tb(self) -> list[Cycle]:
         tb_edges = [
             (u, v, conn)
             for u, v, conn in self._graph.weighted_edge_list()
             if conn.is_thunderbolt()
         ]
 
-        tb_graph: rx.PyDiGraph[NodeInfo, Connection] = rx.PyDiGraph()
+        tb_graph: rx.PyDiGraph[NodeId, SocketConnection] = rx.PyDiGraph()
         tb_graph.add_nodes_from(self._graph.nodes())
 
         for u, v, conn in tb_edges:
-            tb_graph.add_edge(u, v, conn)
+            if isinstance(conn, SocketConnection):
+                tb_graph.add_edge(u, v, conn)
 
         cycle_idxs = rx.simple_cycles(tb_graph)
-        cycles: list[list[NodeInfo]] = []
+        cycles: list[Cycle] = []
         for cycle_idx in cycle_idxs:
-            cycle = [tb_graph[idx] for idx in cycle_idx]
+            cycle = Cycle(node_ids=[tb_graph[idx] for idx in cycle_idx])
             cycles.append(cycle)
 
         return cycles
 
-    def get_subgraph_from_nodes(self, nodes: list[NodeInfo]) -> "Topology":
-        node_idxs = [node.node_id for node in nodes]
-        rx_idxs = [self._node_id_to_rx_id_map[idx] for idx in node_idxs]
+    def get_subgraph_from_nodes(self, node_ids: list[NodeId]) -> "Topology":
         topology = Topology()
-        for rx_idx in rx_idxs:
-            topology.add_node(self._graph[rx_idx])
+        for node_id in node_ids:
+            topology.add_node(node_id)
         for connection in self.list_connections():
-            if (
-                connection.local_node_id in node_idxs
-                and connection.send_back_node_id in node_idxs
-            ):
+            if connection.source in node_ids and connection.sink in node_ids:
                 topology.add_connection(connection)
         return topology
 
-    def is_thunderbolt_cycle(self, cycle: list[NodeInfo]) -> bool:
-        node_idxs = [node.node_id for node in cycle]
-        rx_idxs = [self._node_id_to_rx_id_map[idx] for idx in node_idxs]
+    def is_thunderbolt_cycle(self, cycle: Cycle) -> bool:
+        node_idxs = [node for node in cycle]
+        rx_idxs = [self._vertex_indices[idx] for idx in node_idxs]
         for rid in rx_idxs:
             for neighbor_rid in self._graph.neighbors(rid):
                 if neighbor_rid not in rx_idxs:

src/exo/shared/types/api.py

@@ -11,10 +11,21 @@ from exo.shared.types.worker.instances import Instance, InstanceId, InstanceMeta
 from exo.shared.types.worker.shards import Sharding
 
 FinishReason = Literal[
-    "stop", "length", "tool_calls", "content_filter", "function_call"
+    "stop", "length", "tool_calls", "content_filter", "function_call", "error"
 ]
 
 
+class ErrorInfo(BaseModel):
+    message: str
+    type: str
+    param: str | None = None
+    code: int
+
+
+class ErrorResponse(BaseModel):
+    error: ErrorInfo
+
+
 class ModelListModel(BaseModel):
     id: str
     object: str = "model"

src/exo/shared/types/chunks.py

@@ -22,6 +22,7 @@ class TokenChunk(BaseChunk):
     token_id: int
     finish_reason: FinishReason | None = None
     stats: GenerationStats | None = None
+    error_message: str | None = None
 
 
 class ImageChunk(BaseChunk):

src/exo/shared/types/events.py

@@ -2,14 +2,14 @@ from datetime import datetime
 
 from pydantic import Field
 
-from exo.shared.topology import Connection, NodePerformanceProfile
+from exo.shared.topology import Connection
 from exo.shared.types.chunks import GenerationChunk
 from exo.shared.types.common import CommandId, Id, NodeId, SessionId
-from exo.shared.types.profiling import MemoryPerformanceProfile
 from exo.shared.types.tasks import Task, TaskId, TaskStatus
 from exo.shared.types.worker.downloads import DownloadProgress
 from exo.shared.types.worker.instances import Instance, InstanceId
 from exo.shared.types.worker.runners import RunnerId, RunnerStatus
+from exo.utils.info_gatherer.info_gatherer import GatheredInfo
 from exo.utils.pydantic_ext import CamelCaseModel, TaggedModel
 
 
@@ -76,25 +76,15 @@ class RunnerDeleted(BaseEvent):
     runner_id: RunnerId
 
 
-# TODO
-class NodeCreated(BaseEvent):
-    node_id: NodeId
-
-
 class NodeTimedOut(BaseEvent):
     node_id: NodeId
 
 
-class NodePerformanceMeasured(BaseEvent):
+# TODO: bikeshed this name
+class NodeGatheredInfo(BaseEvent):
     node_id: NodeId
     when: str  # this is a manually cast datetime overrode by the master when the event is indexed, rather than the local time on the device
-    node_profile: NodePerformanceProfile
-
-
-class NodeMemoryMeasured(BaseEvent):
-    node_id: NodeId
-    when: str  # this is a manually cast datetime overrode by the master when the event is indexed, rather than the local time on the device
-    memory: MemoryPerformanceProfile
+    info: GatheredInfo
 
 
 class NodeDownloadProgress(BaseEvent):
@@ -107,11 +97,11 @@ class ChunkGenerated(BaseEvent):
 
 
 class TopologyEdgeCreated(BaseEvent):
-    edge: Connection
+    conn: Connection
 
 
 class TopologyEdgeDeleted(BaseEvent):
-    edge: Connection
+    conn: Connection
 
 
 Event = (
@@ -125,10 +115,8 @@ Event = (
     | InstanceDeleted
     | RunnerStatusUpdated
     | RunnerDeleted
-    | NodeCreated
     | NodeTimedOut
-    | NodePerformanceMeasured
-    | NodeMemoryMeasured
+    | NodeGatheredInfo
     | NodeDownloadProgress
     | ChunkGenerated
     | TopologyEdgeCreated

src/exo/shared/types/multiaddr.py

@@ -1,10 +1,11 @@
 import re
 from typing import ClassVar
 
-from pydantic import BaseModel, computed_field, field_validator
+from pydantic import BaseModel, ConfigDict, computed_field, field_validator
 
 
 class Multiaddr(BaseModel):
+    model_config = ConfigDict(frozen=True)
     address: str
 
     PATTERNS: ClassVar[list[str]] = [

src/exo/shared/types/profiling.py

@@ -1,12 +1,14 @@
+from collections.abc import Sequence
 from typing import Self
 
 import psutil
 
 from exo.shared.types.memory import Memory
+from exo.shared.types.thunderbolt import ThunderboltIdentifier
 from exo.utils.pydantic_ext import CamelCaseModel
 
 
-class MemoryPerformanceProfile(CamelCaseModel):
+class MemoryUsage(CamelCaseModel):
     ram_total: Memory
     ram_available: Memory
     swap_total: Memory
@@ -44,7 +46,6 @@ class SystemPerformanceProfile(CamelCaseModel):
     sys_power: float = 0.0
     pcpu_usage: float = 0.0
     ecpu_usage: float = 0.0
-    ane_power: float = 0.0
 
 
 class NetworkInterfaceInfo(CamelCaseModel):
@@ -53,15 +54,12 @@ class NetworkInterfaceInfo(CamelCaseModel):
 
 
 class NodePerformanceProfile(CamelCaseModel):
-    model_id: str
-    chip_id: str
-    friendly_name: str
-    memory: MemoryPerformanceProfile
-    network_interfaces: list[NetworkInterfaceInfo] = []
-    system: SystemPerformanceProfile
-
-
-class ConnectionProfile(CamelCaseModel):
-    throughput: float
-    latency: float
-    jitter: float
+    model_id: str = "Unknown"
+    chip_id: str = "Unknown"
+    friendly_name: str = "Unknown"
+    memory: MemoryUsage = MemoryUsage.from_bytes(
+        ram_total=0, ram_available=0, swap_total=0, swap_available=0
+    )
+    network_interfaces: Sequence[NetworkInterfaceInfo] = []
+    tb_interfaces: Sequence[ThunderboltIdentifier] = []
+    system: SystemPerformanceProfile = SystemPerformanceProfile()

src/exo/shared/types/thunderbolt.py

@@ -0,0 +1,81 @@
+import anyio
+from pydantic import BaseModel, Field
+
+from exo.utils.pydantic_ext import CamelCaseModel
+
+
+class ThunderboltConnection(CamelCaseModel):
+    source_uuid: str
+    sink_uuid: str
+
+
+class ThunderboltIdentifier(CamelCaseModel):
+    rdma_interface: str
+    domain_uuid: str
+
+
+## Intentionally minimal, only collecting data we care about - there's a lot more
+
+
+class _ReceptacleTag(BaseModel, extra="ignore"):
+    receptacle_id_key: str | None = None
+
+
+class _ConnectivityItem(BaseModel, extra="ignore"):
+    domain_uuid_key: str | None = None
+
+
+class ThunderboltConnectivityData(BaseModel, extra="ignore"):
+    domain_uuid_key: str | None = None
+    items: list[_ConnectivityItem] | None = Field(None, alias="_items")
+    receptacle_1_tag: _ReceptacleTag | None = None
+
+    def ident(self, ifaces: dict[str, str]) -> ThunderboltIdentifier | None:
+        if (
+            self.domain_uuid_key is None
+            or self.receptacle_1_tag is None
+            or self.receptacle_1_tag.receptacle_id_key is None
+        ):
+            return
+        tag = f"Thunderbolt {self.receptacle_1_tag.receptacle_id_key}"
+        assert tag in ifaces  # doesn't need to be an assertion but im confident
+        # if tag not in ifaces: return None
+        iface = f"rdma_{ifaces[tag]}"
+        return ThunderboltIdentifier(
+            rdma_interface=iface, domain_uuid=self.domain_uuid_key
+        )
+
+    def conn(self) -> ThunderboltConnection | None:
+        if self.domain_uuid_key is None or self.items is None:
+            return
+
+        sink_key = next(
+            (
+                item.domain_uuid_key
+                for item in self.items
+                if item.domain_uuid_key is not None
+            ),
+            None,
+        )
+        if sink_key is None:
+            return None
+
+        return ThunderboltConnection(
+            source_uuid=self.domain_uuid_key, sink_uuid=sink_key
+        )
+
+
+class ThunderboltConnectivity(BaseModel, extra="ignore"):
+    SPThunderboltDataType: list[ThunderboltConnectivityData] = []
+
+    @classmethod
+    async def gather(cls) -> list[ThunderboltConnectivityData] | None:
+        proc = await anyio.run_process(
+            ["system_profiler", "SPThunderboltDataType", "-json"], check=False
+        )
+        if proc.returncode != 0:
+            return None
+        # Saving you from PascalCase while avoiding too much pydantic
+        return ThunderboltConnectivity.model_validate_json(
+            proc.stdout
+        ).SPThunderboltDataType

src/exo/shared/types/topology.py

@@ -1,37 +1,41 @@
+from collections.abc import Iterator
+from dataclasses import dataclass
+
 from exo.shared.types.common import NodeId
 from exo.shared.types.multiaddr import Multiaddr
-from exo.shared.types.profiling import ConnectionProfile, NodePerformanceProfile
-from exo.utils.pydantic_ext import CamelCaseModel
+from exo.utils.pydantic_ext import FrozenModel
 
 
-class NodeInfo(CamelCaseModel):
-    node_id: NodeId
-    node_profile: NodePerformanceProfile | None = None
+@dataclass(frozen=True)
+class Cycle:
+    node_ids: list[NodeId]
+
+    def __len__(self) -> int:
+        return self.node_ids.__len__()
+
+    def __iter__(self) -> Iterator[NodeId]:
+        return self.node_ids.__iter__()
 
 
-class Connection(CamelCaseModel):
-    local_node_id: NodeId
-    send_back_node_id: NodeId
-    send_back_multiaddr: Multiaddr
-    connection_profile: ConnectionProfile | None = None
-
-    def __hash__(self) -> int:
-        return hash(
-            (
-                self.local_node_id,
-                self.send_back_node_id,
-                self.send_back_multiaddr.address,
-            )
-        )
-
-    def __eq__(self, other: object) -> bool:
-        if not isinstance(other, Connection):
-            raise ValueError("Cannot compare Connection with non-Connection")
-        return (
-            self.local_node_id == other.local_node_id
-            and self.send_back_node_id == other.send_back_node_id
-            and self.send_back_multiaddr == other.send_back_multiaddr
-        )
+class RDMAConnection(FrozenModel):
+    source_rdma_iface: str
+    sink_rdma_iface: str
 
     def is_thunderbolt(self) -> bool:
-        return str(self.send_back_multiaddr.ipv4_address).startswith("169.254")
+        return True
+
+
+class SocketConnection(FrozenModel):
+    sink_multiaddr: Multiaddr
+
+    def __hash__(self):
+        return hash(self.sink_multiaddr.ip_address)
+
+    def is_thunderbolt(self) -> bool:
+        return str(self.sink_multiaddr.ipv4_address).startswith("169.254")
+
+
+class Connection(FrozenModel):
+    source: NodeId
+    sink: NodeId
+    edge: RDMAConnection | SocketConnection

src/exo/shared/types/worker/downloads.py

@@ -28,7 +28,7 @@ class DownloadPending(BaseDownloadProgress):
 
 
 class DownloadCompleted(BaseDownloadProgress):
-    pass
+    total_bytes: Memory
 
 
 class DownloadFailed(BaseDownloadProgress):

src/exo/shared/types/worker/instances.py

@@ -30,7 +30,7 @@ class MlxRingInstance(BaseInstance):
 
 
 class MlxJacclInstance(BaseInstance):
-    ibv_devices: list[list[str | None]]
+    jaccl_devices: list[list[str | None]]
     jaccl_coordinators: dict[NodeId, str]
 
 

src/exo/shared/types/worker/resource_monitor.py

@@ -1,43 +0,0 @@
-import asyncio
-from abc import ABC, abstractmethod
-from collections.abc import Coroutine
-from typing import Callable
-
-from exo.shared.types.profiling import (
-    MemoryPerformanceProfile,
-    SystemPerformanceProfile,
-)
-
-
-class ResourceCollector(ABC):
-    @abstractmethod
-    async def collect(self) -> SystemPerformanceProfile | MemoryPerformanceProfile: ...
-
-
-class SystemResourceCollector(ResourceCollector):
-    async def collect(self) -> SystemPerformanceProfile: ...
-
-
-class MemoryResourceCollector(ResourceCollector):
-    async def collect(self) -> MemoryPerformanceProfile: ...
-
-
-class ResourceMonitor:
-    data_collectors: list[ResourceCollector]
-    effect_handlers: set[
-        Callable[[SystemPerformanceProfile | MemoryPerformanceProfile], None]
-    ]
-
-    async def _collect(
-        self,
-    ) -> list[SystemPerformanceProfile | MemoryPerformanceProfile]:
-        tasks: list[
-            Coroutine[None, None, SystemPerformanceProfile | MemoryPerformanceProfile]
-        ] = [collector.collect() for collector in self.data_collectors]
-        return await asyncio.gather(*tasks)
-
-    async def collect(self) -> None:
-        profiles = await self._collect()
-        for profile in profiles:
-            for effect_handler in self.effect_handlers:
-                effect_handler(profile)

src/exo/utils/info_gatherer/info_gatherer.py

@@ -0,0 +1,235 @@
+import os
+import shutil
+import sys
+import tomllib
+from collections.abc import Sequence
+from dataclasses import dataclass, field
+from subprocess import CalledProcessError
+from typing import Self, cast
+
+import anyio
+from anyio import create_task_group, open_process
+from anyio.abc import TaskGroup
+from anyio.streams.buffered import BufferedByteReceiveStream
+from anyio.streams.text import TextReceiveStream
+from loguru import logger
+
+from exo.shared.constants import EXO_CONFIG_FILE
+from exo.shared.types.memory import Memory
+from exo.shared.types.profiling import (
+    MemoryUsage,
+    NetworkInterfaceInfo,
+)
+from exo.shared.types.thunderbolt import (
+    ThunderboltConnection,
+    ThunderboltConnectivity,
+    ThunderboltIdentifier,
+)
+from exo.utils.channels import Sender
+from exo.utils.pydantic_ext import TaggedModel
+
+from .macmon import MacmonMetrics
+from .system_info import get_friendly_name, get_model_and_chip, get_network_interfaces
+
+IS_DARWIN = sys.platform == "darwin"
+
+
+class StaticNodeInformation(TaggedModel):
+    """Node information that should NEVER change, to be gathered once at startup"""
+
+    model: str
+    chip: str
+
+    @classmethod
+    async def gather(cls) -> Self:
+        model, chip = await get_model_and_chip()
+        return cls(model=model, chip=chip)
+
+
+class NodeNetworkInterfaces(TaggedModel):
+    ifaces: Sequence[NetworkInterfaceInfo]
+
+
+class MacThunderboltIdentifiers(TaggedModel):
+    idents: Sequence[ThunderboltIdentifier]
+
+
+class MacThunderboltConnections(TaggedModel):
+    conns: Sequence[ThunderboltConnection]
+
+
+class NodeConfig(TaggedModel):
+    """Node configuration from EXO_CONFIG_FILE, reloaded from the file only at startup. Other changes should come in through the API and propagate from there"""
+
+    @classmethod
+    async def gather(cls) -> Self | None:
+        cfg_file = anyio.Path(EXO_CONFIG_FILE)
+        await cfg_file.touch(exist_ok=True)
+        async with await cfg_file.open("rb") as f:
+            try:
+                contents = (await f.read()).decode("utf-8")
+                data = tomllib.loads(contents)
+                return cls.model_validate(data)
+            except (tomllib.TOMLDecodeError, UnicodeDecodeError):
+                logger.warning("Invalid config file, skipping...")
+                return None
+
+
+class MiscData(TaggedModel):
+    """Node information that may slowly change that doesn't fall into the other categories"""
+
+    friendly_name: str
+
+    @classmethod
+    async def gather(cls) -> Self:
+        return cls(friendly_name=await get_friendly_name())
+
+
+async def _gather_iface_map() -> dict[str, str] | None:
+    proc = await anyio.run_process(
+        ["networksetup", "-listallhardwareports"], check=False
+    )
+    if proc.returncode != 0:
+        return None
+
+    ports: dict[str, str] = {}
+    port = ""
+    for line in proc.stdout.decode("utf-8").split("\n"):
+        if line.startswith("Hardware Port:"):
+            port = line.split(": ")[1]
+        elif line.startswith("Device:"):
+            ports[port] = line.split(": ")[1]
+            port = ""
+    if "" in ports:
+        del ports[""]
+    return ports
+
+
+GatheredInfo = (
+    MacmonMetrics
+    | MemoryUsage
+    | NodeNetworkInterfaces
+    | MacThunderboltIdentifiers
+    | MacThunderboltConnections
+    | NodeConfig
+    | MiscData
+    | StaticNodeInformation
+)
+
+
+@dataclass
+class InfoGatherer:
+    info_sender: Sender[GatheredInfo]
+    interface_watcher_interval: float | None = 10
+    misc_poll_interval: float | None = 60
+    system_profiler_interval: float | None = 5 if IS_DARWIN else None
+    memory_poll_rate: float | None = None if IS_DARWIN else 1
+    macmon_interval: float | None = 1 if IS_DARWIN else None
+    _tg: TaskGroup = field(init=False, default_factory=create_task_group)
+
+    async def run(self):
+        async with self._tg as tg:
+            if IS_DARWIN:
+                if (macmon_path := shutil.which("macmon")) is not None:
+                    tg.start_soon(self._monitor_macmon, macmon_path)
+                tg.start_soon(self._monitor_system_profiler_thunderbolt_data)
+            tg.start_soon(self._watch_system_info)
+            tg.start_soon(self._monitor_memory_usage)
+            tg.start_soon(self._monitor_misc)
+
+            nc = await NodeConfig.gather()
+            if nc is not None:
+                await self.info_sender.send(nc)
+            sni = await StaticNodeInformation.gather()
+            await self.info_sender.send(sni)
+
+    def shutdown(self):
+        self._tg.cancel_scope.cancel()
+
+    async def _monitor_misc(self):
+        if self.misc_poll_interval is None:
+            return
+        prev = await MiscData.gather()
+        await self.info_sender.send(prev)
+        while True:
+            curr = await MiscData.gather()
+            if prev != curr:
+                prev = curr
+                await self.info_sender.send(curr)
+            await anyio.sleep(self.misc_poll_interval)
+
+    async def _monitor_system_profiler_thunderbolt_data(self):
+        if self.system_profiler_interval is None:
+            return
+        iface_map = await _gather_iface_map()
+        if iface_map is None:
+            return
+
+        old_idents = []
+        while True:
+            data = await ThunderboltConnectivity.gather()
+            assert data is not None
+
+            idents = [it for i in data if (it := i.ident(iface_map)) is not None]
+            if idents != old_idents:
+                await self.info_sender.send(MacThunderboltIdentifiers(idents=idents))
+            old_idents = idents
+
+            conns = [it for i in data if (it := i.conn()) is not None]
+            await self.info_sender.send(MacThunderboltConnections(conns=conns))
+
+            await anyio.sleep(self.system_profiler_interval)
+
+    async def _monitor_memory_usage(self):
+        override_memory_env = os.getenv("OVERRIDE_MEMORY_MB")
+        override_memory: int | None = (
+            Memory.from_mb(int(override_memory_env)).in_bytes
+            if override_memory_env
+            else None
+        )
+        if self.memory_poll_rate is None:
+            return
+        while True:
+            await self.info_sender.send(
+                MemoryUsage.from_psutil(override_memory=override_memory)
+            )
+            await anyio.sleep(self.memory_poll_rate)
+
+    async def _watch_system_info(self):
+        if self.interface_watcher_interval is None:
+            return
+        old_nics = []
+        while True:
+            nics = get_network_interfaces()
+            if nics != old_nics:
+                old_nics = nics
+                await self.info_sender.send(NodeNetworkInterfaces(ifaces=nics))
+            await anyio.sleep(self.interface_watcher_interval)
+
+    async def _monitor_macmon(self, macmon_path: str):
+        if self.macmon_interval is None:
+            return
+        # macmon pipe --interval [interval in ms]
+        try:
+            async with await open_process(
+                [macmon_path, "pipe", "--interval", str(self.macmon_interval * 1000)]
+            ) as p:
+                if not p.stdout:
+                    logger.critical("MacMon closed stdout")
+                    return
+                async for text in TextReceiveStream(
+                    BufferedByteReceiveStream(p.stdout)
+                ):
+                    await self.info_sender.send(MacmonMetrics.from_raw_json(text))
+        except CalledProcessError as e:
+            stderr_msg = "no stderr"
+            stderr_output = cast(bytes | str | None, e.stderr)
+            if stderr_output is not None:
+                stderr_msg = (
+                    stderr_output.decode()
+                    if isinstance(stderr_output, bytes)
+                    else str(stderr_output)
+                )
+            logger.warning(
+                f"MacMon failed with return code {e.returncode}: {stderr_msg}"
+            )

src/exo/utils/info_gatherer/macmon.py

@@ -0,0 +1,70 @@
+from typing import Self
+
+from pydantic import BaseModel
+
+from exo.shared.types.profiling import MemoryUsage, SystemPerformanceProfile
+from exo.utils.pydantic_ext import TaggedModel
+
+
+class _TempMetrics(BaseModel, extra="ignore"):
+    """Temperature-related metrics returned by macmon."""
+
+    cpu_temp_avg: float
+    gpu_temp_avg: float
+
+
+class _MemoryMetrics(BaseModel, extra="ignore"):
+    """Memory-related metrics returned by macmon."""
+
+    ram_total: int
+    ram_usage: int
+    swap_total: int
+    swap_usage: int
+
+
+class RawMacmonMetrics(BaseModel, extra="ignore"):
+    """Complete set of metrics returned by macmon.
+
+    Unknown fields are ignored for forward-compatibility.
+    """
+
+    timestamp: str  # ignored
+    temp: _TempMetrics
+    memory: _MemoryMetrics
+    ecpu_usage: tuple[int, float]  # freq mhz, usage %
+    pcpu_usage: tuple[int, float]  # freq mhz, usage %
+    gpu_usage: tuple[int, float]  # freq mhz, usage %
+    all_power: float
+    ane_power: float
+    cpu_power: float
+    gpu_power: float
+    gpu_ram_power: float
+    ram_power: float
+    sys_power: float
+
+
+class MacmonMetrics(TaggedModel):
+    system_profile: SystemPerformanceProfile
+    memory: MemoryUsage
+
+    @classmethod
+    def from_raw(cls, raw: RawMacmonMetrics) -> Self:
+        return cls(
+            system_profile=SystemPerformanceProfile(
+                gpu_usage=raw.gpu_usage[1],
+                temp=raw.temp.gpu_temp_avg,
+                sys_power=raw.sys_power,
+                pcpu_usage=raw.pcpu_usage[1],
+                ecpu_usage=raw.ecpu_usage[1],
+            ),
+            memory=MemoryUsage.from_bytes(
+                ram_total=raw.memory.ram_total,
+                ram_available=(raw.memory.ram_total - raw.memory.ram_usage),
+                swap_total=raw.memory.swap_total,
+                swap_available=(raw.memory.swap_total - raw.memory.swap_usage),
+            ),
+        )
+
+    @classmethod
+    def from_raw_json(cls, json: str) -> Self:
+        return cls.from_raw(RawMacmonMetrics.model_validate_json(json))

src/exo/utils/info_gatherer/net_profile.py

@@ -0,0 +1,114 @@
+from collections.abc import Mapping
+
+import anyio
+import httpx
+from anyio import create_task_group
+from loguru import logger
+
+from exo.shared.topology import Topology
+from exo.shared.types.common import NodeId
+from exo.shared.types.profiling import NodePerformanceProfile
+
+REACHABILITY_ATTEMPTS = 3
+
+
+async def check_reachability(
+    target_ip: str,
+    expected_node_id: NodeId,
+    out: dict[NodeId, set[str]],
+    client: httpx.AsyncClient,
+) -> None:
+    """Check if a node is reachable at the given IP and verify its identity."""
+    if ":" in target_ip:
+        # TODO: use real IpAddress types
+        url = f"http://[{target_ip}]:52415/node_id"
+    else:
+        url = f"http://{target_ip}:52415/node_id"
+
+    remote_node_id = None
+    last_error = None
+
+    for _ in range(REACHABILITY_ATTEMPTS):
+        try:
+            r = await client.get(url)
+            if r.status_code != 200:
+                await anyio.sleep(1)
+                continue
+
+            body = r.text.strip().strip('"')
+            if not body:
+                await anyio.sleep(1)
+                continue
+
+            remote_node_id = NodeId(body)
+            break
+
+        # expected failure cases
+        except (
+            httpx.TimeoutException,
+            httpx.NetworkError,
+        ):
+            await anyio.sleep(1)
+
+        # other failures should be logged on last attempt
+        except httpx.HTTPError as e:
+            last_error = e
+            await anyio.sleep(1)
+
+    if last_error is not None:
+        logger.warning(
+            f"connect error {type(last_error).__name__} from {target_ip} after {REACHABILITY_ATTEMPTS} attempts; treating as down"
+        )
+
+    if remote_node_id is None:
+        return
+
+    if remote_node_id != expected_node_id:
+        logger.warning(
+            f"Discovered node with unexpected node_id; "
+            f"ip={target_ip}, expected_node_id={expected_node_id}, "
+            f"remote_node_id={remote_node_id}"
+        )
+        return
+
+    if remote_node_id not in out:
+        out[remote_node_id] = set()
+    out[remote_node_id].add(target_ip)
+
+
+async def check_reachable(
+    topology: Topology,
+    self_node_id: NodeId,
+    node_profiles: Mapping[NodeId, NodePerformanceProfile],
+) -> dict[NodeId, set[str]]:
+    """Check which nodes are reachable and return their IPs."""
+
+    reachable: dict[NodeId, set[str]] = {}
+
+    # these are intentionally httpx's defaults so we can tune them later
+    timeout = httpx.Timeout(timeout=5.0)
+    limits = httpx.Limits(
+        max_connections=100,
+        max_keepalive_connections=20,
+        keepalive_expiry=5,
+    )
+
+    async with (
+        httpx.AsyncClient(timeout=timeout, limits=limits) as client,
+        create_task_group() as tg,
+    ):
+        for node_id in topology.list_nodes():
+            if node_id not in node_profiles:
+                continue
+            if node_id == self_node_id:
+                continue
+            for iface in node_profiles[node_id].network_interfaces:
+                tg.start_soon(
+                    check_reachability,
+                    iface.ip_address,
+                    node_id,
+                    reachable,
+                    client,
+                )
+
+    return reachable

src/exo/utils/info_gatherer/tests/test_tb_parsing.py

@@ -0,0 +1,24 @@
+import sys
+
+import pytest
+
+from exo.shared.types.thunderbolt import (
+    ThunderboltConnectivity,
+)
+from exo.utils.info_gatherer.info_gatherer import (
+    _gather_iface_map,  # pyright: ignore[reportPrivateUsage]
+)
+
+
+@pytest.mark.anyio
+@pytest.mark.skipif(
+    sys.platform != "darwin", reason="Thunderbolt info can only be gathered on macos"
+)
+async def test_tb_parsing():
+    data = await ThunderboltConnectivity.gather()
+    ifaces = await _gather_iface_map()
+    assert ifaces
+    assert data
+    for datum in data:
+        datum.ident(ifaces)
+        datum.conn()

src/exo/utils/pydantic_ext.py

@@ -19,11 +19,20 @@ class CamelCaseModel(BaseModel):
         alias_generator=to_camel,
         validate_by_name=True,
         extra="forbid",
-        # I want to reenable this ASAP, but it's causing an issue with TaskStatus
         strict=True,
     )
 
 
+class FrozenModel(BaseModel):
+    model_config = ConfigDict(
+        alias_generator=to_camel,
+        validate_by_name=True,
+        extra="forbid",
+        strict=True,
+        frozen=True,
+    )
+
+
 class TaggedModel(CamelCaseModel):
     @model_serializer(mode="wrap")
     def _serialize(self, handler: SerializerFunctionWrapHandler):

src/exo/utils/tests/test_mp_channel.py

@@ -28,9 +28,8 @@ def bar(send: MpSender[str]):
     send.close()
 
 
-# not async, just want the fail_after
 @pytest.mark.anyio
-async def test_channel_setup():
+async def test_channel_ipc():
     with fail_after(0.5):
         s, r = mp_channel[str]()
         p1 = mp.Process(target=foo, args=(r,))

src/exo/worker/download/download_utils.py

@@ -5,6 +5,7 @@ import shutil
 import ssl
 import time
 import traceback
+from collections.abc import Awaitable
 from datetime import timedelta
 from pathlib import Path
 from typing import Callable, Literal
@@ -245,12 +246,15 @@ def create_http_session(
         sock_read_timeout = 1800
         sock_connect_timeout = 60
 
-    ssl_context = ssl.create_default_context(cafile=certifi.where())
+    ssl_context = ssl.create_default_context(
+        cafile=os.getenv("SSL_CERT_FILE") or certifi.where()
+    )
     connector = aiohttp.TCPConnector(ssl=ssl_context)
 
     return aiohttp.ClientSession(
         auto_decompress=auto_decompress,
         connector=connector,
+        proxy=os.getenv("HTTPS_PROXY") or os.getenv("HTTP_PROXY") or None,
         timeout=aiohttp.ClientTimeout(
             total=total_timeout,
             connect=connect_timeout,
@@ -522,7 +526,7 @@ async def download_progress_for_local_path(
 
 async def download_shard(
     shard: ShardMetadata,
-    on_progress: Callable[[ShardMetadata, RepoDownloadProgress], None],
+    on_progress: Callable[[ShardMetadata, RepoDownloadProgress], Awaitable[None]],
     max_parallel_downloads: int = 8,
     skip_download: bool = False,
     allow_patterns: list[str] | None = None,
@@ -563,9 +567,9 @@ async def download_shard(
     )
     file_progress: dict[str, RepoFileDownloadProgress] = {}
 
-    def on_progress_wrapper(
+    async def on_progress_wrapper(
         file: FileListEntry, curr_bytes: int, total_bytes: int, is_renamed: bool
-    ):
+    ) -> None:
         start_time = (
             file_progress[file.path].start_time
             if file.path in file_progress
@@ -601,7 +605,7 @@ async def download_shard(
             else "in_progress",
             start_time=start_time,
         )
-        on_progress(
+        await on_progress(
             shard,
             calculate_repo_progress(
                 shard,
@@ -629,14 +633,21 @@ async def download_shard(
 
     semaphore = asyncio.Semaphore(max_parallel_downloads)
 
-    async def download_with_semaphore(file: FileListEntry):
+    def schedule_progress(
+        file: FileListEntry, curr_bytes: int, total_bytes: int, is_renamed: bool
+    ) -> None:
+        asyncio.create_task(
+            on_progress_wrapper(file, curr_bytes, total_bytes, is_renamed)
+        )
+
+    async def download_with_semaphore(file: FileListEntry) -> None:
         async with semaphore:
             await download_file_with_retry(
                 str(shard.model_meta.model_id),
                 revision,
                 file.path,
                 target_dir,
-                lambda curr_bytes, total_bytes, is_renamed: on_progress_wrapper(
+                lambda curr_bytes, total_bytes, is_renamed: schedule_progress(
                     file, curr_bytes, total_bytes, is_renamed
                 ),
             )
@@ -648,7 +659,7 @@ async def download_shard(
     final_repo_progress = calculate_repo_progress(
         shard, str(shard.model_meta.model_id), revision, file_progress, all_start_time
     )
-    on_progress(shard, final_repo_progress)
+    await on_progress(shard, final_repo_progress)
     if gguf := next((f for f in filtered_file_list if f.path.endswith(".gguf")), None):
         return target_dir / gguf.path, final_repo_progress
     else:

src/exo/worker/download/impl_shard_downloader.py

@@ -1,4 +1,5 @@
 import asyncio
+from collections.abc import Awaitable
 from pathlib import Path
 from typing import AsyncIterator, Callable
 
@@ -48,7 +49,8 @@ class SingletonShardDownloader(ShardDownloader):
         self.active_downloads: dict[ShardMetadata, asyncio.Task[Path]] = {}
 
     def on_progress(
-        self, callback: Callable[[ShardMetadata, RepoDownloadProgress], None]
+        self,
+        callback: Callable[[ShardMetadata, RepoDownloadProgress], Awaitable[None]],
     ) -> None:
         self.shard_downloader.on_progress(callback)
 
@@ -83,7 +85,8 @@ class CachedShardDownloader(ShardDownloader):
         self.cache: dict[tuple[str, ShardMetadata], Path] = {}
 
     def on_progress(
-        self, callback: Callable[[ShardMetadata, RepoDownloadProgress], None]
+        self,
+        callback: Callable[[ShardMetadata, RepoDownloadProgress], Awaitable[None]],
     ) -> None:
         self.shard_downloader.on_progress(callback)
 
@@ -113,17 +116,18 @@ class ResumableShardDownloader(ShardDownloader):
     def __init__(self, max_parallel_downloads: int = 8):
         self.max_parallel_downloads = max_parallel_downloads
         self.on_progress_callbacks: list[
-            Callable[[ShardMetadata, RepoDownloadProgress], None]
+            Callable[[ShardMetadata, RepoDownloadProgress], Awaitable[None]]
         ] = []
 
-    def on_progress_wrapper(
+    async def on_progress_wrapper(
         self, shard: ShardMetadata, progress: RepoDownloadProgress
     ) -> None:
         for callback in self.on_progress_callbacks:
-            callback(shard, progress)
+            await callback(shard, progress)
 
     def on_progress(
-        self, callback: Callable[[ShardMetadata, RepoDownloadProgress], None]
+        self,
+        callback: Callable[[ShardMetadata, RepoDownloadProgress], Awaitable[None]],
     ) -> None:
         self.on_progress_callbacks.append(callback)
 

src/exo/worker/download/shard_downloader.py

@@ -1,4 +1,5 @@
 from abc import ABC, abstractmethod
+from collections.abc import Awaitable
 from copy import copy
 from datetime import timedelta
 from pathlib import Path
@@ -31,7 +32,8 @@ class ShardDownloader(ABC):
 
     @abstractmethod
     def on_progress(
-        self, callback: Callable[[ShardMetadata, RepoDownloadProgress], None]
+        self,
+        callback: Callable[[ShardMetadata, RepoDownloadProgress], Awaitable[None]],
     ) -> None:
         pass
 
@@ -59,7 +61,8 @@ class NoopShardDownloader(ShardDownloader):
         return Path("/tmp/noop_shard")
 
     def on_progress(
-        self, callback: Callable[[ShardMetadata, RepoDownloadProgress], None]
+        self,
+        callback: Callable[[ShardMetadata, RepoDownloadProgress], Awaitable[None]],
     ) -> None:
         pass
 

src/exo/worker/engines/mlx/auto_parallel.py

@@ -1,7 +1,10 @@
+import os
+import threading
 from abc import ABC, abstractmethod
+from collections.abc import Callable
 from functools import partial
 from inspect import signature
-from typing import TYPE_CHECKING, Callable, Protocol, cast
+from typing import TYPE_CHECKING, Any, Protocol, cast
 
 import mlx.core as mx
 import mlx.nn as nn
@@ -10,18 +13,58 @@ from mlx.nn.layers.distributed import (
     shard_linear,
     sum_gradients,
 )
-from mlx_lm.models.cache import (
-    _BaseCache,  # pyright: ignore[reportPrivateUsage]
-)
 from mlx_lm.models.deepseek_v3 import DeepseekV3MLP
 from mlx_lm.models.deepseek_v3 import Model as DeepseekV3Model
+from mlx_lm.models.deepseek_v32 import DeepseekV32MLP
+from mlx_lm.models.deepseek_v32 import Model as DeepseekV32Model
+from mlx_lm.models.glm4_moe import Model as Glm4MoeModel
+from mlx_lm.models.glm4_moe import MoE
+from mlx_lm.models.gpt_oss import GptOssMoeModel
+from mlx_lm.models.gpt_oss import Model as GptOssModel
 from mlx_lm.models.llama import Model as LlamaModel
+from mlx_lm.models.minimax import Model as MiniMaxModel
+from mlx_lm.models.ministral3 import Model as Ministral3Model
 from mlx_lm.models.qwen3_moe import Model as Qwen3MoeModel
 from mlx_lm.models.qwen3_moe import Qwen3MoeSparseMoeBlock
+from mlx_lm.models.qwen3_next import Model as Qwen3NextModel
+from mlx_lm.models.qwen3_next import Qwen3NextSparseMoeBlock
 
-from exo.shared.types.worker.shards import (
-    PipelineShardMetadata,
-)
+from exo.shared.logging import logger
+from exo.shared.types.worker.shards import PipelineShardMetadata
+
+TimeoutCallback = Callable[[], None]
+
+
+def eval_with_timeout(
+    mlx_item: Any,  # pyright: ignore[reportAny]
+    timeout_seconds: float = 60.0,
+    on_timeout: TimeoutCallback | None = None,
+) -> None:
+    """Evaluate MLX item with a hard timeout.
+
+    If on_timeout callback is provided, it will be called before terminating
+    the process. This allows the runner to send a failure event before exit.
+    """
+    completed = threading.Event()
+
+    def watchdog() -> None:
+        if not completed.wait(timeout=timeout_seconds):
+            logger.error(
+                f"mlx_item evaluation timed out after {timeout_seconds:.0f}s. "
+                "This may indicate an issue with FAST_SYNCH and tensor parallel sharding. "
+                "Terminating process."
+            )
+            if on_timeout is not None:
+                on_timeout()
+            os._exit(1)
+
+    watchdog_thread = threading.Thread(target=watchdog, daemon=True)
+    watchdog_thread.start()
+
+    try:
+        mx.eval(mlx_item)  # pyright: ignore[reportAny]
+    finally:
+        completed.set()
 
 
 class _LayerCallable(Protocol):
@@ -40,9 +83,11 @@ class CustomMlxLayer(nn.Module):
 
     def __init__(self, original_layer: _LayerCallable):
         super().__init__()
-        # Set twice to avoid __setattr__ recursion
         object.__setattr__(self, "_original_layer", original_layer)
-        self.original_layer: _LayerCallable = original_layer
+
+    @property
+    def original_layer(self) -> _LayerCallable:
+        return cast(_LayerCallable, object.__getattribute__(self, "_original_layer"))
 
     # Calls __getattr__ for any attributes not found on nn.Module (e.g. use_sliding)
     if not TYPE_CHECKING:
@@ -52,7 +97,7 @@ class CustomMlxLayer(nn.Module):
                 return super().__getattr__(name)
             except AttributeError:
                 original_layer = object.__getattribute__(self, "_original_layer")
-                return object.__getattribute__(original_layer, name)
+                return getattr(original_layer, name)
 
 
 class PipelineFirstLayer(CustomMlxLayer):
@@ -91,8 +136,6 @@ class PipelineLastLayer(CustomMlxLayer):
             x, *args, **kwargs
         ).arguments.get("cache", None)
 
-        assert cache is None or issubclass(type(cache), _BaseCache)  # type: ignore
-
         output: mx.array = self.original_layer(x, *args, **kwargs)
 
         if self.r != self.s - 1:
@@ -100,10 +143,8 @@ class PipelineLastLayer(CustomMlxLayer):
                 output, (self.r + 1) % self.s, group=self.group
             )
             if cache is not None:
-                # This change happened upstream - check out mlx github somewhere??
                 cache.keys = mx.depends(cache.keys, output)  # type: ignore[reportUnknownMemberType]
 
-        output = mx.distributed.all_gather(output, group=self.group)[-output.shape[0] :]
         return output
 
 
@@ -132,28 +173,30 @@ def _get_layers(inner_model_instance: nn.Module) -> list[_LayerCallable]:
     return layers
 
 
-def _set_layers(model: nn.Module, layers: list[_LayerCallable]) -> None:
-    inner_model_instance = _inner_model(model)
-    if hasattr(inner_model_instance, "layers"):
-        inner_model_instance.layers = layers
+class _IdentityModule(nn.Module):
+    """Identity module that returns input unchanged. Used to skip computation."""
 
-        # Update DeepSeek V3 specific parameters when layers are shrunk
-        if isinstance(model, DeepseekV3Model) and hasattr(
-            inner_model_instance, "num_layers"
-        ):
-            inner_model_instance.start_idx = 0
-            inner_model_instance.end_idx = len(layers)
-            inner_model_instance.num_layers = len(layers)
-    elif hasattr(inner_model_instance, "h"):
-        inner_model_instance.h = layers
-    else:
-        raise ValueError("Model must have either a 'layers' or 'h' attribute")
+    def __call__(self, x: mx.array, *args: object, **kwargs: object) -> mx.array:
+        return x
+
+
+class _IdentityLmHead(nn.Module):
+    """Identity lm_head that returns zeros. Used for non-final pipeline ranks."""
+
+    def __init__(self, vocab_size: int, dtype: mx.Dtype = mx.float16):
+        super().__init__()
+        self.vocab_size = vocab_size
+        self.dtype = dtype
+
+    def __call__(self, x: mx.array) -> mx.array:
+        # Return zeros with correct shape (batch, seq, vocab_size)
+        return mx.zeros((*x.shape[:-1], self.vocab_size), dtype=self.dtype)
 
 
 def pipeline_auto_parallel(
     model: nn.Module,
     group: mx.distributed.Group,
-    model_shard_meta: PipelineShardMetadata,
+    model_shard_meta: PipelineShardMetadata
 ) -> nn.Module:
     """
     Automatically parallelize a model across multiple devices.
@@ -165,13 +208,13 @@ def pipeline_auto_parallel(
     """
     inner_model_instance: nn.Module = _inner_model(model)
 
-    # Handle both model.layers and model.h cases
-    layers: list[_LayerCallable] = _get_layers(inner_model_instance)
+    layers = _get_layers(inner_model_instance)
 
     start_layer, end_layer = model_shard_meta.start_layer, model_shard_meta.end_layer
     device_rank, world_size = model_shard_meta.device_rank, model_shard_meta.world_size
 
     layers = layers[start_layer:end_layer]
+
     layers[0] = PipelineFirstLayer(layers[0], device_rank, group=group)
     layers[-1] = PipelineLastLayer(
         layers[-1],
@@ -180,18 +223,97 @@ def pipeline_auto_parallel(
         group=group,
     )
 
+    if isinstance(inner_model_instance, GptOssMoeModel):
+        inner_model_instance.layer_types = inner_model_instance.layer_types[  # type: ignore
+            start_layer:end_layer
+        ]
+        # We can assume the model has at least one layer thanks to placement.
+        # If a layer type doesn't exist, we can set it to 0.
+        inner_model_instance.swa_idx = (
+            0
+            if "sliding_attention" not in inner_model_instance.layer_types  # type: ignore
+            else inner_model_instance.layer_types.index(  # type: ignore
+                "sliding_attention"
+            )
+        )
+        inner_model_instance.ga_idx = (
+            0
+            if "full_attention" not in inner_model_instance.layer_types  # type: ignore
+            else inner_model_instance.layer_types.index(  # type: ignore
+                "full_attention"
+            )
+        )
+
     _set_layers(model, layers)
 
     assert isinstance(layers, list), (
         "Expected a list of layers after auto-parallel initialisation"
     )
 
+    return patch_pipeline_model(model, group)
+
+
+def patch_pipeline_model[T](model: T, group: mx.distributed.Group) -> T:
+    # Patch __call__ on the model's class
+    cls = model.__class__
+    original_call = cls.__call__  # type :ignore
+    call_signature = signature(original_call)  # type :ignore
+
+    def patched_call(
+        self: T,
+        *args: object,
+        **kwargs: object,
+    ) -> mx.array:
+        logits: mx.array = original_call(self, *args, **kwargs)  # type: ignore
+        cache = call_signature.bind_partial(self, *args, **kwargs).arguments.get(
+            "cache", None
+        )
+
+        # Add dependency to last cache entry to ensure distributed ops are evaluated
+        if cache is not None:
+            cache[-1].state = mx.depends(cache[-1].state, logits)  # type: ignore
+
+        logits = mx.distributed.all_gather(logits, group=group)[
+            -logits.shape[0] :
+        ]  # type :ignore
+
+        return logits
+
+    cls.__call__ = patched_call
+    return model
+
+
+def patch_tensor_model[T](model: T) -> T:
+    """Patch model's __call__ to ensure distributed ops sync during inference."""
+    cls = model.__class__
+    original_call = cls.__call__
+    call_signature = signature(original_call)
+
+    def patched_call(
+        self: T,
+        *args: object,
+        **kwargs: object,
+    ) -> mx.array:
+        logits: mx.array = original_call(self, *args, **kwargs)  # pyright: ignore[reportAny]
+        cache = call_signature.bind_partial(self, *args, **kwargs).arguments.get(
+            "cache", None
+        )
+
+        # Add dependency to last cache entry to ensure distributed ops are evaluated
+        if cache is not None and len(cache) > 0:  # pyright: ignore[reportAny]
+            cache[-1].state = mx.depends(cache[-1].state, logits)  # pyright: ignore[reportAny,reportUnknownMemberType]
+
+        return logits
+
+    cls.__call__ = patched_call
     return model
 
 
 def tensor_auto_parallel(
     model: nn.Module,
     group: mx.distributed.Group,
+    timeout_seconds: float = 60.0,
+    on_timeout: TimeoutCallback | None = None,
 ) -> nn.Module:
     all_to_sharded_linear = partial(
         shard_linear,
@@ -204,18 +326,44 @@ def tensor_auto_parallel(
         group=group,
     )
 
+    segments: int = 1
+
+    def _all_to_sharded(path: str, weight: mx.array):
+        if path.endswith("bias"):
+            logger.info(f"Sharding bias for {path} - all to sharded")
+            return weight.ndim - 1, segments
+        return max(weight.ndim - 2, 0), segments
+
     all_to_sharded_linear_in_place = partial(
         shard_inplace,
-        sharding="all-to-sharded",
-        group=group,
-    )
-    sharded_to_all_linear_in_place = partial(
-        shard_inplace,
-        sharding="sharded-to-all",
+        sharding=_all_to_sharded,  # type: ignore
         group=group,
     )
 
-    if isinstance(model, LlamaModel):
+    n = group.size()
+
+    def _sharded_to_all(path: str, weight: mx.array):
+        if path.endswith("bias"):
+            logger.info(f"Sharding bias for {path} - sharded to all")
+            weight /= n
+            return None
+        return -1, segments
+
+    sharded_to_all_linear_in_place = partial(
+        shard_inplace,
+        sharding=_sharded_to_all,  # type: ignore
+        group=group,
+    )
+
+    if hasattr(model, "shard"):
+        try:
+            model.shard(group)  # type: ignore
+            return patch_tensor_model(model)
+        except (AttributeError, TypeError, NameError):
+            pass
+
+    if isinstance(model, (LlamaModel, Ministral3Model)):
+        logger.warning("shouldn't be hit - upstream sharding exists")
         tensor_parallel_sharding_strategy = LlamaShardingStrategy(
             group,
             all_to_sharded_linear,
@@ -223,7 +371,8 @@ def tensor_auto_parallel(
             all_to_sharded_linear_in_place,
             sharded_to_all_linear_in_place,
         )
-    elif isinstance(model, DeepseekV3Model):
+    elif isinstance(model, (DeepseekV3Model, DeepseekV32Model)):
+        logger.warning("shouldn't be hit - upstream sharding exists")
         tensor_parallel_sharding_strategy = DeepSeekShardingStrategy(
             group,
             all_to_sharded_linear,
@@ -231,7 +380,15 @@ def tensor_auto_parallel(
             all_to_sharded_linear_in_place,
             sharded_to_all_linear_in_place,
         )
-    elif isinstance(model, Qwen3MoeModel):
+    elif isinstance(model, MiniMaxModel):
+        tensor_parallel_sharding_strategy = MiniMaxShardingStrategy(
+            group,
+            all_to_sharded_linear,
+            sharded_to_all_linear,
+            all_to_sharded_linear_in_place,
+            sharded_to_all_linear_in_place,
+        )
+    elif isinstance(model, (Qwen3MoeModel, Glm4MoeModel, Qwen3NextModel)):
         tensor_parallel_sharding_strategy = QwenShardingStrategy(
             group,
             all_to_sharded_linear,
@@ -239,10 +396,22 @@ def tensor_auto_parallel(
             all_to_sharded_linear_in_place,
             sharded_to_all_linear_in_place,
         )
+    elif isinstance(model, GptOssModel):
+        tensor_parallel_sharding_strategy = GptOssShardingStrategy(
+            group,
+            all_to_sharded_linear,
+            sharded_to_all_linear,
+            all_to_sharded_linear_in_place,
+            sharded_to_all_linear_in_place,
+        )
+
     else:
         raise ValueError(f"Unsupported model type: {type(model)}")
 
-    return tensor_parallel_sharding_strategy.shard_model(model)
+    model = tensor_parallel_sharding_strategy.shard_model(
+        model, timeout_seconds, on_timeout
+    )
+    return patch_tensor_model(model)
 
 
 class TensorParallelShardingStrategy(ABC):
@@ -262,13 +431,27 @@ class TensorParallelShardingStrategy(ABC):
         self.N = group.size()
 
     @abstractmethod
-    def shard_model(self, model: nn.Module) -> nn.Module: ...
+    def shard_model(
+        self,
+        model: nn.Module,
+        timeout_seconds: float,
+        on_timeout: TimeoutCallback | None,
+    ) -> nn.Module: ...
 
 
 class LlamaShardingStrategy(TensorParallelShardingStrategy):
-    def shard_model(self, model: nn.Module) -> nn.Module:
+    def shard_model(
+        self,
+        model: nn.Module,
+        timeout_seconds: float,
+        on_timeout: TimeoutCallback | None,
+    ) -> nn.Module:
         model = cast(LlamaModel, model)
         for layer in model.layers:
+            # Force load weights before sharding to avoid FAST_SYNCH deadlock
+            eval_with_timeout(
+                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
+            )
             layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
             layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
             layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
@@ -284,13 +467,46 @@ class LlamaShardingStrategy(TensorParallelShardingStrategy):
         return model
 
 
+def _set_layers(model: nn.Module, layers: list[_LayerCallable]) -> None:
+    inner_model_instance = _inner_model(model)
+    if hasattr(inner_model_instance, "layers"):
+        inner_model_instance.layers = layers
+
+        # Update DeepSeek V3 specific parameters when layers are shrunk
+        if isinstance(
+            model, (DeepseekV3Model, DeepseekV32Model, Glm4MoeModel)
+        ) and hasattr(inner_model_instance, "num_layers"):
+            logger.info(
+                f"Setting num_layers to {len(layers)} for model {model.model.__class__.__name__}"
+            )
+            inner_model_instance.start_idx = 0
+            inner_model_instance.end_idx = len(layers)
+            inner_model_instance.num_layers = len(layers)
+        elif isinstance(model, Qwen3MoeModel):
+            logger.info(
+                f"Setting num_hidden_layers to {len(layers)} for model {model.model.__class__.__name__}"
+            )
+            inner_model_instance.num_hidden_layers = len(layers)
+    elif hasattr(inner_model_instance, "h"):
+        inner_model_instance.h = layers
+    else:
+        raise ValueError("Model must have either a 'layers' or 'h' attribute")
+
+
 class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
-    def shard_model(self, model: nn.Module) -> nn.Module:
+    def shard_model(
+        self,
+        model: nn.Module,
+        timeout_seconds: float,
+        on_timeout: TimeoutCallback | None,
+    ) -> nn.Module:
         model = cast(DeepseekV3Model, model)
         for layer in model.layers:
+            eval_with_timeout(
+                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
+            )
             # Shard the self attention
-            if layer.self_attn.q_lora_rank is None:  # pyright: ignore[reportUnnecessaryComparison]
-                # Unfortunately, q_lora_rank can be None despite typing hints.
+            if layer.self_attn.q_lora_rank is None:
                 layer.self_attn.q_proj = self.all_to_sharded_linear(
                     layer.self_attn.q_proj
                 )
@@ -305,7 +521,7 @@ class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
             layer.self_attn.num_heads //= self.N
 
             # Shard the MLP
-            if isinstance(layer.mlp, DeepseekV3MLP):
+            if isinstance(layer.mlp, (DeepseekV3MLP, DeepseekV32MLP)):
                 layer.mlp.gate_proj = self.all_to_sharded_linear(layer.mlp.gate_proj)
                 layer.mlp.down_proj = self.sharded_to_all_linear(layer.mlp.down_proj)
                 layer.mlp.up_proj = self.all_to_sharded_linear(layer.mlp.up_proj)
@@ -339,10 +555,55 @@ class ShardedDeepseekV3MoE(CustomMlxLayer):
         return y
 
 
+class MiniMaxShardingStrategy(TensorParallelShardingStrategy):
+    def shard_model(
+        self,
+        model: nn.Module,
+        timeout_seconds: float,
+        on_timeout: TimeoutCallback | None,
+    ) -> nn.Module:
+        model = cast(MiniMaxModel, model)
+        for layer in model.layers:
+            eval_with_timeout(
+                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
+            )
+            # Shard the self attention
+            layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
+            layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
+            layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
+            layer.self_attn.o_proj = self.sharded_to_all_linear(layer.self_attn.o_proj)
+            layer.self_attn.num_attention_heads //= self.N
+            layer.self_attn.num_key_value_heads //= self.N
+
+            # Shard the MoE. Shard in place since the MoE should be responsible
+            # for aggregating the results.
+            self.all_to_sharded_linear_in_place(
+                layer.block_sparse_moe.switch_mlp.gate_proj
+            )
+            self.sharded_to_all_linear_in_place(
+                layer.block_sparse_moe.switch_mlp.down_proj
+            )
+            self.all_to_sharded_linear_in_place(
+                layer.block_sparse_moe.switch_mlp.up_proj
+            )
+            layer.block_sparse_moe = ShardedQwenMoE(layer.block_sparse_moe)  # pyright: ignore[reportAttributeAccessIssue, reportArgumentType]
+            layer.block_sparse_moe.sharding_group = self.group
+
+        return model
+
+
 class QwenShardingStrategy(TensorParallelShardingStrategy):
-    def shard_model(self, model: nn.Module) -> nn.Module:
+    def shard_model(
+        self,
+        model: nn.Module,
+        timeout_seconds: float,
+        on_timeout: TimeoutCallback | None,
+    ) -> nn.Module:
         model = cast(Qwen3MoeModel, model)
         for layer in model.layers:
+            eval_with_timeout(
+                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
+            )
             # Shard the self attention
             layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
             layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
@@ -353,11 +614,13 @@ class QwenShardingStrategy(TensorParallelShardingStrategy):
 
             # Shard the MoE. Shard in place since the MoE should be responsible
             # for aggregating the results.
-            if isinstance(layer.mlp, Qwen3MoeSparseMoeBlock):
+            if isinstance(
+                layer.mlp, (Qwen3MoeSparseMoeBlock, MoE, Qwen3NextSparseMoeBlock)
+            ):
                 self.all_to_sharded_linear_in_place(layer.mlp.switch_mlp.gate_proj)
                 self.sharded_to_all_linear_in_place(layer.mlp.switch_mlp.down_proj)
                 self.all_to_sharded_linear_in_place(layer.mlp.switch_mlp.up_proj)
-                layer.mlp = ShardedQwenMoE(layer.mlp)  # type: ignore
+                layer.mlp = ShardedQwenMoE(layer.mlp)  # pyright: ignore[reportAttributeAccessIssue, reportArgumentType]
                 layer.mlp.sharding_group = self.group
 
             # Shard the MLP
@@ -381,3 +644,58 @@ class ShardedQwenMoE(CustomMlxLayer):
         if self.sharding_group is not None:
             y = mx.distributed.all_sum(y, group=self.sharding_group)
         return y
+
+
+class GptOssShardingStrategy(TensorParallelShardingStrategy):
+    def shard_model(
+        self,
+        model: nn.Module,
+        timeout_seconds: float,
+        on_timeout: TimeoutCallback | None,
+    ) -> nn.Module:
+        model = cast(GptOssMoeModel, model)
+
+        for layer in model.layers:
+            eval_with_timeout(
+                layer.parameters(), timeout_seconds / len(model.layers), on_timeout
+            )
+            layer.self_attn.q_proj = self.all_to_sharded_linear(layer.self_attn.q_proj)
+            layer.self_attn.k_proj = self.all_to_sharded_linear(layer.self_attn.k_proj)
+            layer.self_attn.v_proj = self.all_to_sharded_linear(layer.self_attn.v_proj)
+            layer.self_attn.o_proj = self.sharded_to_all_linear(layer.self_attn.o_proj)
+
+            layer.self_attn.num_attention_heads //= self.N
+            layer.self_attn.num_key_value_heads //= self.N
+            layer.self_attn.num_key_value_groups = (
+                layer.self_attn.num_attention_heads
+                // layer.self_attn.num_key_value_heads
+            )
+
+            layer.self_attn.sinks = layer.self_attn.sinks[
+                layer.self_attn.num_attention_heads
+                * self.group.rank() : layer.self_attn.num_attention_heads
+                * (self.group.rank() + 1)
+            ]
+
+            self.all_to_sharded_linear_in_place(layer.mlp.experts.gate_proj)
+            self.sharded_to_all_linear_in_place(layer.mlp.experts.down_proj)
+            self.all_to_sharded_linear_in_place(layer.mlp.experts.up_proj)
+
+            layer.mlp = ShardedGptOssMoE(layer.mlp)  # type: ignore
+            layer.mlp.sharding_group = self.group
+
+        return model
+
+
+class ShardedGptOssMoE(CustomMlxLayer):
+    def __init__(self, layer: nn.Module):
+        super().__init__(layer)
+        self.sharding_group: mx.distributed.Group | None = None
+
+    def __call__(self, x: mx.array) -> mx.array:
+        if self.sharding_group is not None:
+            x = sum_gradients(self.sharding_group)(x)
+        y = self.original_layer(x)
+        if self.sharding_group is not None:
+            y = mx.distributed.all_sum(y, group=self.sharding_group)
+        return y

src/exo/worker/engines/mlx/generator/generate.py

@@ -3,6 +3,7 @@ from typing import Any, Callable, Generator, cast, get_args
 import mlx.core as mx
 from mlx_lm import stream_generate
 from mlx_lm.models.cache import KVCache
+from mlx_lm.sample_utils import make_sampler
 from mlx_lm.tokenizer_utils import TokenizerWrapper
 
 # from exo.engines.mlx.cache import KVPrefixCache
@@ -47,7 +48,6 @@ def maybe_quantize_kv_cache(
 def warmup_inference(
     model: Model,
     tokenizer: TokenizerWrapper,
-    sampler: Callable[[mx.array], mx.array],
 ) -> int:
     content = "Prompt to warm up the inference engine. Repeat this."
 
@@ -70,6 +70,9 @@ def warmup_inference(
         model=model,
     )
 
+    # Use a default sampler for warmup
+    sampler = make_sampler(temp=0.7)
+
     logger.info("Generating warmup tokens")
     for _r in stream_generate(
         model=model,
@@ -115,7 +118,6 @@ def eos_ids_from_tokenizer(tokenizer: TokenizerWrapper) -> list[int]:
 def mlx_generate(
     model: Model,
     tokenizer: TokenizerWrapper,
-    sampler: Callable[[mx.array], mx.array],
     task: ChatCompletionTaskParams,
 ) -> Generator[GenerationResponse]:
     # Ensure that generation stats only contains peak memory for this generation
@@ -125,6 +127,9 @@ def mlx_generate(
     # Currently we support chat-completion tasks only.
     logger.info(f"task_params: {task}")
 
+    if task.seed is not None:
+        mx.random.seed(task.seed)
+
     prompt = apply_chat_template(
         tokenizer=tokenizer,
         chat_task_data=task,
@@ -138,6 +143,11 @@ def mlx_generate(
         eos_ids = eos_ids_from_tokenizer(tokenizer)
         logits_processors = [ban_token_ids(eos_ids)]
 
+    sampler = make_sampler(
+        temp=task.temperature if task.temperature is not None else 0.7,
+        top_p=task.top_p if task.top_p is not None else 1.0,
+    )
+
     max_tokens = task.max_tokens or MAX_TOKENS
     for out in stream_generate(
         model=model,

src/exo/worker/engines/mlx/utils_mlx.py

@@ -1,13 +1,26 @@
 import json
 import os
 import resource
+import sys
 import time
 from pathlib import Path
-from typing import Any, Callable, cast
+from typing import Any, cast
+
+# Monkey-patch for transformers 5.x compatibility
+# Kimi's tokenization_kimi.py imports bytes_to_unicode from the old location
+# which was moved in transformers 5.0.0rc2
+try:
+    import transformers.models.gpt2.tokenization_gpt2 as gpt2_tokenization
+    from transformers.convert_slow_tokenizer import bytes_to_unicode
+
+    if not hasattr(gpt2_tokenization, "bytes_to_unicode"):
+        gpt2_tokenization.bytes_to_unicode = bytes_to_unicode  # type: ignore[attr-defined]
+except ImportError:
+    pass  # transformers < 5.0 or bytes_to_unicode not available
 
 from mlx_lm.models.cache import KVCache, QuantizedKVCache, RotatingKVCache
 from mlx_lm.models.deepseek_v3 import DeepseekV3Model
-from mlx_lm.sample_utils import make_sampler
+from mlx_lm.models.gpt_oss import Model as GptOssModel
 from mlx_lm.tokenizer_utils import TokenizerWrapper
 
 from exo.worker.engines.mlx.constants import (
@@ -19,7 +32,7 @@ from exo.worker.engines.mlx.constants import (
 try:
     from mlx_lm.tokenizer_utils import load_tokenizer
 except ImportError:
-    from mlx_lm.tokenizer_utils import load as load_tokenizer  # type: ignore
+    from mlx_lm.tokenizer_utils import load as load_tokenizer
 import contextlib
 
 import mlx.core as mx
@@ -44,6 +57,8 @@ from exo.shared.types.worker.shards import (
 from exo.worker.download.download_utils import build_model_path
 from exo.worker.engines.mlx import Model
 from exo.worker.engines.mlx.auto_parallel import (
+    TimeoutCallback,
+    eval_with_timeout,
     pipeline_auto_parallel,
     tensor_auto_parallel,
 )
@@ -69,6 +84,10 @@ def get_weights_size(model_shard_meta: ShardMetadata) -> Memory:
     )
 
 
+class ModelLoadingTimeoutError(Exception):
+    pass
+
+
 def mx_barrier(group: Group | None = None):
     mx.eval(
         mx.distributed.all_sum(
@@ -132,20 +151,26 @@ def mlx_distributed_init(
                 group = mx.distributed.init(backend="ring", strict=True)
 
             case MlxJacclInstance(
-                ibv_devices=ibv_devices, jaccl_coordinators=jaccl_coordinators
+                jaccl_devices=jaccl_devices, jaccl_coordinators=jaccl_coordinators
             ):
+                assert all(
+                    jaccl_devices[i][i] is None for i in range(len(jaccl_devices))
+                )
                 # Use RDMA connectivity matrix
                 coordination_file = (
                     f"./hosts_{bound_instance.instance.instance_id}_{rank}.json"
                 )
-                ibv_devices_json = json.dumps(ibv_devices)
+                jaccl_devices_json = json.dumps(jaccl_devices)
 
                 with open(coordination_file, "w") as f:
-                    _ = f.write(ibv_devices_json)
+                    _ = f.write(jaccl_devices_json)
 
                 jaccl_coordinator = jaccl_coordinators[bound_instance.bound_node_id]
 
-                logger.info(f"rank {rank} MLX_IBV_DEVICES: {ibv_devices_json}")
+                # TODO: update once upstream fixes
+                logger.info(
+                    f"rank {rank} MLX_IBV_DEVICES: {coordination_file} with devices: {jaccl_devices_json}"
+                )
                 logger.info(f"rank {rank} MLX_JACCL_COORDINATOR: {jaccl_coordinator}")
                 os.environ["MLX_IBV_DEVICES"] = coordination_file
                 os.environ["MLX_RANK"] = str(rank)
@@ -175,12 +200,10 @@ def initialize_mlx(
 
 
 def load_mlx_items(
-    bound_instance: BoundInstance, group: Group | None
-) -> tuple[Model, TokenizerWrapper, Callable[[mx.array], mx.array]]:
-    # TODO: pass temperature
-    sampler: Callable[[mx.array], mx.array] = make_sampler(temp=0.7)
-    logger.info("Created a sampler")
-
+    bound_instance: BoundInstance,
+    group: Group | None,
+    on_timeout: TimeoutCallback | None = None,
+) -> tuple[Model, TokenizerWrapper]:
     if group is None:
         logger.info(f"Single device used for {bound_instance.instance}")
         model_path = build_model_path(bound_instance.bound_shard.model_meta.model_id)
@@ -193,7 +216,9 @@ def load_mlx_items(
     else:
         logger.info("Starting distributed init")
         start_time = time.perf_counter()
-        model, tokenizer = shard_and_load(bound_instance.bound_shard, group=group)
+        model, tokenizer = shard_and_load(
+            bound_instance.bound_shard, group=group, on_timeout=on_timeout
+        )
         end_time = time.perf_counter()
         logger.info(
             f"Time taken to shard and load model: {(end_time - start_time):.2f}s"
@@ -201,12 +226,13 @@ def load_mlx_items(
 
     set_wired_limit_for_model(get_weights_size(bound_instance.bound_shard))
 
-    return cast(Model, model), tokenizer, sampler
+    return cast(Model, model), tokenizer
 
 
 def shard_and_load(
     shard_metadata: ShardMetadata,
     group: Group,
+    on_timeout: TimeoutCallback | None = None,
 ) -> tuple[nn.Module, TokenizerWrapper]:
     model_path = build_model_path(shard_metadata.model_meta.model_id)
 
@@ -235,18 +261,30 @@ def shard_and_load(
 
     logger.info(f"Group size: {group.size()}, group rank: {group.rank()}")
 
+    # Estimate timeout based on model size
+    base_timeout = float(os.environ.get("EXO_MODEL_LOAD_TIMEOUT", "60"))
+    model_size_gb = get_weights_size(shard_metadata).in_bytes / (1024**3)
+    timeout_seconds = base_timeout + model_size_gb / 5
+    logger.info(
+        f"Evaluating model parameters with timeout of {timeout_seconds:.0f}s "
+        f"(model size: {model_size_gb:.1f}GB)"
+    )
+
     match shard_metadata:
         case TensorShardMetadata():
             logger.info(f"loading model from {model_path} with tensor parallelism")
-            model = tensor_auto_parallel(model, group)
+            model = tensor_auto_parallel(model, group, timeout_seconds, on_timeout)
         case PipelineShardMetadata():
             logger.info(f"loading model from {model_path} with pipeline parallelism")
-            model = pipeline_auto_parallel(model, group, shard_metadata)
+            model = pipeline_auto_parallel(
+                model, group, shard_metadata
+            )
+            # Skip eval for pipeline parallel to avoid fast synch issues
+            mx_barrier(group)
+            return model, tokenizer
 
-    mx.eval(model.parameters())
-
-    # TODO: Do we need this?
-    mx.eval(model)
+    # Eager eval for tensor parallel (ranks have same operations on sharded data)
+    eval_with_timeout(model.parameters(), timeout_seconds, on_timeout)
 
     logger.debug("SHARDED")
     logger.debug(model)
@@ -257,26 +295,70 @@ def shard_and_load(
     return model, tokenizer
 
 
-def get_tokenizer(model_path: Path, shard_metadata: ShardMetadata):
-    # TODO: Let's move away from this custom logic to mlx_lm.load()
-    if "kimi-k2" in shard_metadata.model_meta.model_id.lower():
-        eos_token_ids = [163586]
+def get_tokenizer(model_path: Path, shard_metadata: ShardMetadata) -> TokenizerWrapper:
+    """Load tokenizer for a model shard. Delegates to load_tokenizer_for_model_id."""
+    return load_tokenizer_for_model_id(shard_metadata.model_meta.model_id, model_path)
 
-    elif "glm" in shard_metadata.model_meta.model_id.lower():
-        eos_token_ids = [151336, 151329, 151338]
 
-    else:
-        eos_token_ids = None
+def get_eos_token_ids_for_model(model_id: str) -> list[int] | None:
+    """
+    Get the EOS token IDs for a model based on its ID.
 
-    tokenizer = cast(
-        TokenizerWrapper,
-        load_tokenizer(
-            model_path,
-            tokenizer_config_extra={"trust_remote_code": TRUST_REMOTE_CODE},
-            eos_token_ids=eos_token_ids,
-        ),
+    Some models require explicit EOS token configuration that isn't in their
+    tokenizer config. This function returns the known EOS token IDs for such models.
+
+    Args:
+        model_id: The HuggingFace model ID
+
+    Returns:
+        List of EOS token IDs, or None if the model uses standard tokenizer config
+    """
+    model_id_lower = model_id.lower()
+    if "kimi-k2" in model_id_lower:
+        return [163586]
+    elif "glm" in model_id_lower:
+        return [151336, 151329, 151338]
+    return None
+
+
+def load_tokenizer_for_model_id(model_id: str, model_path: Path) -> TokenizerWrapper:
+    """
+    Load tokenizer for a model given its ID and local path.
+
+    This is the core tokenizer loading logic, handling special cases for different
+    model families (Kimi, GLM, etc.) and transformers 5.x compatibility.
+
+    Args:
+        model_id: The HuggingFace model ID (e.g., "moonshotai/Kimi-K2-Instruct")
+        model_path: Local path where the model/tokenizer files are stored
+
+    Returns:
+        TokenizerWrapper instance configured for the model
+    """
+    model_id_lower = model_id.lower()
+    eos_token_ids = get_eos_token_ids_for_model(model_id)
+
+    # Kimi uses a custom TikTokenTokenizer that transformers 5.x can't load via AutoTokenizer
+    if "kimi-k2" in model_id_lower:
+        sys.path.insert(0, str(model_path))
+        from tokenization_kimi import TikTokenTokenizer  # type: ignore[import-not-found]  # noqa: I001
+
+        hf_tokenizer: Any = TikTokenTokenizer.from_pretrained(model_path)  # pyright: ignore[reportUnknownVariableType,reportUnknownMemberType]
+
+        # Patch encode to use internal tiktoken model directly
+        # transformers 5.x has a bug in the encode->pad path for slow tokenizers
+        def _patched_encode(text: str, **_kwargs: object) -> list[int]:
+            # Pass allowed_special="all" to handle special tokens like <|im_user|>
+            return list(hf_tokenizer.model.encode(text, allowed_special="all"))  # pyright: ignore[reportUnknownMemberType,reportUnknownArgumentType]
+
+        hf_tokenizer.encode = _patched_encode
+        return TokenizerWrapper(hf_tokenizer, eos_token_ids=eos_token_ids)
+
+    tokenizer = load_tokenizer(
+        model_path,
+        tokenizer_config_extra={"trust_remote_code": TRUST_REMOTE_CODE},
+        eos_token_ids=eos_token_ids,
     )
-    assert isinstance(tokenizer, TokenizerWrapper)
 
     return tokenizer
 
@@ -289,15 +371,15 @@ def apply_chat_template(
     messages = chat_task_data.messages
 
     formatted_messages: list[dict[str, Any]] = []
-    for _, message in enumerate(messages):
+    for message in messages:
         if isinstance(message.content, ChatCompletionMessageText):
             message.content = message.content.text
         if isinstance(message.content, list):
-            if len(message.content) != 1:
-                logger.warning("Received malformed prompt")
+            if len(message.content) == 0:
+                logger.warning("Received prompt with no content, skipping")
                 continue
 
-            message.content = message.content[0].text
+            message.content = "\n".join(c.text for c in message.content).strip()
         if message.content is None and message.thinking is None:
             continue
 
@@ -306,13 +388,16 @@ def apply_chat_template(
             {k: v for k, v in message.model_dump().items() if v is not None}  # type: ignore
         )
 
-    prompt: str = tokenizer.apply_chat_template(  # type: ignore
+    prompt: str = tokenizer.apply_chat_template(
         formatted_messages,
         tokenize=False,
         add_generation_prompt=True,
+        tools=chat_task_data.tools,
     )
 
-    return prompt  # type: ignore
+    logger.info(prompt)
+
+    return prompt
 
 
 class NullKVCache(KVCache):
@@ -343,6 +428,11 @@ def make_kv_cache(
 ) -> list[KVCache | RotatingKVCache | QuantizedKVCache]:
     assert hasattr(model, "layers")
 
+    # TODO: Do this for all models
+    if hasattr(model, "make_cache") and isinstance(model, GptOssModel):
+        logger.info("Using MLX LM's make cache")
+        return model.make_cache()  # type: ignore
+
     if max_kv_size is None:
         if KV_CACHE_BITS is None:
             logger.info("Using default KV cache")

src/exo/worker/main.py

@@ -16,8 +16,7 @@ from exo.shared.types.events import (
     ForwarderEvent,
     IndexedEvent,
     NodeDownloadProgress,
-    NodeMemoryMeasured,
-    NodePerformanceMeasured,
+    NodeGatheredInfo,
     TaskCreated,
     TaskStatusUpdated,
     TopologyEdgeCreated,
@@ -25,7 +24,6 @@ from exo.shared.types.events import (
 )
 from exo.shared.types.models import ModelId
 from exo.shared.types.multiaddr import Multiaddr
-from exo.shared.types.profiling import MemoryPerformanceProfile, NodePerformanceProfile
 from exo.shared.types.state import State
 from exo.shared.types.tasks import (
     CreateRunner,
@@ -34,7 +32,7 @@ from exo.shared.types.tasks import (
     Task,
     TaskStatus,
 )
-from exo.shared.types.topology import Connection
+from exo.shared.types.topology import Connection, SocketConnection
 from exo.shared.types.worker.downloads import (
     DownloadCompleted,
     DownloadOngoing,
@@ -45,14 +43,14 @@ from exo.shared.types.worker.runners import RunnerId
 from exo.shared.types.worker.shards import ShardMetadata
 from exo.utils.channels import Receiver, Sender, channel
 from exo.utils.event_buffer import OrderedBuffer
+from exo.utils.info_gatherer.info_gatherer import GatheredInfo, InfoGatherer
+from exo.utils.info_gatherer.net_profile import check_reachable
 from exo.worker.download.download_utils import (
     map_repo_download_progress_to_download_progress_data,
 )
 from exo.worker.download.shard_downloader import RepoDownloadProgress, ShardDownloader
 from exo.worker.plan import plan
 from exo.worker.runner.runner_supervisor import RunnerSupervisor
-from exo.worker.utils import start_polling_memory_metrics, start_polling_node_metrics
-from exo.worker.utils.net_profile import check_reachable
 
 
 class Worker:
@@ -86,7 +84,7 @@ class Worker:
         self.state: State = State()
         self.download_status: dict[ModelId, DownloadProgress] = {}
         self.runners: dict[RunnerId, RunnerSupervisor] = {}
-        self._tg: TaskGroup | None = None
+        self._tg: TaskGroup = create_task_group()
 
         self._nack_cancel_scope: CancelScope | None = None
         self._nack_attempts: int = 0
@@ -98,37 +96,13 @@ class Worker:
     async def run(self):
         logger.info("Starting Worker")
 
-        # TODO: CLEANUP HEADER
-        async def resource_monitor_callback(
-            node_performance_profile: NodePerformanceProfile,
-        ) -> None:
-            await self.event_sender.send(
-                NodePerformanceMeasured(
-                    node_id=self.node_id,
-                    node_profile=node_performance_profile,
-                    when=str(datetime.now(tz=timezone.utc)),
-                ),
-            )
+        info_send, info_recv = channel[GatheredInfo]()
+        info_gatherer: InfoGatherer = InfoGatherer(info_send)
 
-        async def memory_monitor_callback(
-            memory_profile: MemoryPerformanceProfile,
-        ) -> None:
-            await self.event_sender.send(
-                NodeMemoryMeasured(
-                    node_id=self.node_id,
-                    memory=memory_profile,
-                    when=str(datetime.now(tz=timezone.utc)),
-                )
-            )
-
-        # END CLEANUP
-
-        async with create_task_group() as tg:
-            self._tg = tg
+        async with self._tg as tg:
+            tg.start_soon(info_gatherer.run)
+            tg.start_soon(self._forward_info, info_recv)
             tg.start_soon(self.plan_step)
-            tg.start_soon(start_polling_node_metrics, resource_monitor_callback)
-
-            tg.start_soon(start_polling_memory_metrics, memory_monitor_callback)
             tg.start_soon(self._emit_existing_download_progress)
             tg.start_soon(self._connection_message_event_writer)
             tg.start_soon(self._resend_out_for_delivery)
@@ -142,6 +116,17 @@ class Worker:
         for runner in self.runners.values():
             runner.shutdown()
 
+    async def _forward_info(self, recv: Receiver[GatheredInfo]):
+        with recv as info_stream:
+            async for info in info_stream:
+                await self.event_sender.send(
+                    NodeGatheredInfo(
+                        node_id=self.node_id,
+                        when=str(datetime.now(tz=timezone.utc)),
+                        info=info,
+                    )
+                )
+
     async def _event_applier(self):
         with self.global_event_receiver as events:
             async for f_event in events:
@@ -161,7 +146,6 @@ class Worker:
                     self._nack_cancel_scope is None
                     or self._nack_cancel_scope.cancel_called
                 ):
-                    assert self._tg
                     # Request the next index.
                     self._tg.start_soon(
                         self._nack_request, self.state.last_event_applied_idx + 1
@@ -217,7 +201,9 @@ class Worker:
                     )
                     if initial_progress.status == "complete":
                         progress = DownloadCompleted(
-                            shard_metadata=shard, node_id=self.node_id
+                            shard_metadata=shard,
+                            node_id=self.node_id,
+                            total_bytes=initial_progress.total_bytes,
                         )
                         self.download_status[shard.model_meta.model_id] = progress
                         await self.event_sender.send(
@@ -250,8 +236,7 @@ class Worker:
                     await self.runners[self._task_to_runner_id(task)].start_task(task)
 
     def shutdown(self):
-        if self._tg:
-            self._tg.cancel_scope.cancel()
+        self._tg.cancel_scope.cancel()
 
     def _task_to_runner_id(self, task: Task):
         instance = self.state.instances[task.instance_id]
@@ -268,24 +253,28 @@ class Worker:
         match msg.connection_type:
             case ConnectionMessageType.Connected:
                 return TopologyEdgeCreated(
-                    edge=Connection(
-                        local_node_id=self.node_id,
-                        send_back_node_id=msg.node_id,
-                        send_back_multiaddr=Multiaddr(
-                            address=f"/ip4/{msg.remote_ipv4}/tcp/{msg.remote_tcp_port}"
+                    conn=Connection(
+                        source=self.node_id,
+                        sink=msg.node_id,
+                        edge=SocketConnection(
+                            sink_multiaddr=Multiaddr(
+                                address=f"/ip4/{msg.remote_ipv4}/tcp/{msg.remote_tcp_port}"
+                            ),
                         ),
-                    )
+                    ),
                 )
 
             case ConnectionMessageType.Disconnected:
                 return TopologyEdgeDeleted(
-                    edge=Connection(
-                        local_node_id=self.node_id,
-                        send_back_node_id=msg.node_id,
-                        send_back_multiaddr=Multiaddr(
-                            address=f"/ip4/{msg.remote_ipv4}/tcp/{msg.remote_tcp_port}"
+                    conn=Connection(
+                        source=self.node_id,
+                        sink=msg.node_id,
+                        edge=SocketConnection(
+                            sink_multiaddr=Multiaddr(
+                                address=f"/ip4/{msg.remote_ipv4}/tcp/{msg.remote_tcp_port}"
+                            ),
                         ),
-                    )
+                    ),
                 )
 
     async def _nack_request(self, since_idx: int) -> None:
@@ -334,7 +323,6 @@ class Worker:
             event_sender=self.event_sender.clone(),
         )
         self.runners[task.bound_instance.bound_runner_id] = runner
-        assert self._tg
         self._tg.start_soon(runner.run)
         return runner
 
@@ -357,20 +345,22 @@ class Worker:
         last_progress_time = 0.0
         throttle_interval_secs = 1.0
 
-        # TODO: i hate callbacks
-        def download_progress_callback(
+        async def download_progress_callback(
             shard: ShardMetadata, progress: RepoDownloadProgress
         ) -> None:
             nonlocal self
             nonlocal last_progress_time
             if progress.status == "complete":
-                status = DownloadCompleted(shard_metadata=shard, node_id=self.node_id)
+                status = DownloadCompleted(
+                    shard_metadata=shard,
+                    node_id=self.node_id,
+                    total_bytes=progress.total_bytes,
+                )
                 self.download_status[shard.model_meta.model_id] = status
-                # Footgun!
-                self.event_sender.send_nowait(
+                await self.event_sender.send(
                     NodeDownloadProgress(download_progress=status)
                 )
-                self.event_sender.send_nowait(
+                await self.event_sender.send(
                     TaskStatusUpdated(
                         task_id=task.task_id, task_status=TaskStatus.Complete
                     )
@@ -387,13 +377,12 @@ class Worker:
                     ),
                 )
                 self.download_status[shard.model_meta.model_id] = status
-                self.event_sender.send_nowait(
+                await self.event_sender.send(
                     NodeDownloadProgress(download_progress=status)
                 )
                 last_progress_time = current_time()
 
         self.shard_downloader.on_progress(download_progress_callback)
-        assert self._tg
         self._tg.start_soon(self.shard_downloader.ensure_shard, task.shard_metadata)
 
     async def _forward_events(self) -> None:
@@ -414,9 +403,14 @@ class Worker:
 
     async def _poll_connection_updates(self):
         while True:
-            # TODO: EdgeDeleted
-            edges = set(self.state.topology.list_connections())
-            conns = await check_reachable(self.state.topology, self.node_id)
+            edges = set(
+                conn.edge for conn in self.state.topology.out_edges(self.node_id)
+            )
+            conns = await check_reachable(
+                self.state.topology,
+                self.node_id,
+                self.state.node_profiles,
+            )
             for nid in conns:
                 for ip in conns[nid]:
                     if "127.0.0.1" in ip or "localhost" in ip:
@@ -424,26 +418,33 @@ class Worker:
                             f"Loopback connection should not happen: {ip=} for {nid=}"
                         )
 
-                    edge = Connection(
-                        local_node_id=self.node_id,
-                        send_back_node_id=nid,
+                    edge = SocketConnection(
                         # nonsense multiaddr
-                        send_back_multiaddr=Multiaddr(address=f"/ip4/{ip}/tcp/52415")
+                        sink_multiaddr=Multiaddr(address=f"/ip4/{ip}/tcp/52415")
                         if "." in ip
                         # nonsense multiaddr
                         else Multiaddr(address=f"/ip6/{ip}/tcp/52415"),
                     )
                     if edge not in edges:
                         logger.debug(f"ping discovered {edge=}")
-                        await self.event_sender.send(TopologyEdgeCreated(edge=edge))
+                        await self.event_sender.send(
+                            TopologyEdgeCreated(
+                                conn=Connection(
+                                    source=self.node_id, sink=nid, edge=edge
+                                )
+                            )
+                        )
 
-            for nid, conn in self.state.topology.out_edges(self.node_id):
+            for conn in self.state.topology.out_edges(self.node_id):
+                if not isinstance(conn.edge, SocketConnection):
+                    continue
                 if (
-                    nid not in conns
-                    or conn.send_back_multiaddr.ip_address not in conns.get(nid, set())
+                    conn.sink not in conns
+                    or conn.edge.sink_multiaddr.ip_address
+                    not in conns.get(conn.sink, set())
                 ):
                     logger.debug(f"ping failed to discover {conn=}")
-                    await self.event_sender.send(TopologyEdgeDeleted(edge=conn))
+                    await self.event_sender.send(TopologyEdgeDeleted(conn=conn))
 
             await anyio.sleep(10)
 
@@ -457,7 +458,9 @@ class Worker:
                 ) in self.shard_downloader.get_shard_download_status():
                     if progress.status == "complete":
                         status = DownloadCompleted(
-                            node_id=self.node_id, shard_metadata=progress.shard
+                            node_id=self.node_id,
+                            shard_metadata=progress.shard,
+                            total_bytes=progress.total_bytes,
                         )
                     elif progress.status in ["in_progress", "not_started"]:
                         if progress.downloaded_bytes_this_session.in_bytes == 0:

src/exo/worker/runner/bootstrap.py

@@ -17,15 +17,23 @@ def entrypoint(
     task_receiver: MpReceiver[Task],
     _logger: "loguru.Logger",
 ) -> None:
-    if (
-        isinstance(bound_instance.instance, MlxJacclInstance)
-        and len(bound_instance.instance.ibv_devices) >= 2
+    fast_synch_override = os.environ.get("EXO_FAST_SYNCH")
+    if fast_synch_override == "on" or (
+        fast_synch_override != "off"
+        and (
+            isinstance(bound_instance.instance, MlxJacclInstance)
+            and len(bound_instance.instance.jaccl_devices) >= 2
+        )
     ):
         os.environ["MLX_METAL_FAST_SYNCH"] = "1"
+    else:
+        os.environ["MLX_METAL_FAST_SYNCH"] = "0"
 
     global logger
     logger = _logger
 
+    logger.info(f"Fast synch flag: {os.environ['MLX_METAL_FAST_SYNCH']}")
+
     # Import main after setting global logger - this lets us just import logger from this module
     try:
         from exo.worker.runner.runner import main

src/exo/worker/runner/runner.py

@@ -1,6 +1,15 @@
 import time
+from collections.abc import Generator
+from functools import cache
 
 import mlx.core as mx
+from mlx_lm.models.gpt_oss import Model as GptOssModel
+from openai_harmony import (  # pyright: ignore[reportMissingTypeStubs]
+    HarmonyEncodingName,
+    Role,
+    StreamableParser,
+    load_harmony_encoding,
+)
 
 from exo.shared.types.api import ChatCompletionMessageText
 from exo.shared.types.chunks import TokenChunk
@@ -58,6 +67,7 @@ def main(
         bound_instance.bound_runner_id,
         bound_instance.bound_shard,
     )
+    device_rank = shard_metadata.device_rank
     logger.info("hello from the runner")
     if getattr(shard_metadata, "immediate_exception", False):
         raise Exception("Fake exception - runner failed to spin up.")
@@ -68,7 +78,6 @@ def main(
 
     model = None
     tokenizer = None
-    sampler = None
     group = None
 
     current_status: RunnerStatus = RunnerIdle()
@@ -110,14 +119,26 @@ def main(
                         )
                     )
 
-                    model, tokenizer, sampler = load_mlx_items(bound_instance, group)
+                    def on_model_load_timeout() -> None:
+                        event_sender.send(
+                            RunnerStatusUpdated(
+                                runner_id=runner_id,
+                                runner_status=RunnerFailed(
+                                    error_message="Model loading timed out"
+                                ),
+                            )
+                        )
+                        time.sleep(0.5)
+
+                    model, tokenizer = load_mlx_items(
+                        bound_instance, group, on_timeout=on_model_load_timeout
+                    )
 
                     current_status = RunnerLoaded()
                     logger.info("runner loaded")
                 case StartWarmup() if isinstance(current_status, RunnerLoaded):
                     assert model
                     assert tokenizer
-                    assert sampler
                     current_status = RunnerWarmingUp()
                     logger.info("runner warming up")
                     event_sender.send(
@@ -130,7 +151,6 @@ def main(
                     toks = warmup_inference(
                         model=model,
                         tokenizer=tokenizer,
-                        sampler=sampler,
                         # kv_prefix_cache=kv_prefix_cache,  # supply for warmup-time prefix caching
                     )
                     logger.info(f"warmed up by generating {toks} tokens")
@@ -142,9 +162,6 @@ def main(
                 case ChatCompletion(task_params=task_params, command_id=command_id) if (
                     isinstance(current_status, RunnerReady)
                 ):
-                    assert model
-                    assert tokenizer
-                    assert sampler
                     logger.info(f"received chat request: {str(task)[:500]}")
                     current_status = RunnerRunning()
                     logger.info("runner running")
@@ -153,34 +170,61 @@ def main(
                             runner_id=runner_id, runner_status=current_status
                         )
                     )
+                    assert model
+                    assert tokenizer
                     assert task_params.messages[0].content is not None
-                    _check_for_debug_prompts(task_params.messages[0].content)
 
-                    # Generate responses using the actual MLX generation
-                    for response in mlx_generate(
-                        model=model,
-                        tokenizer=tokenizer,
-                        sampler=sampler,
-                        task=task_params,
-                    ):
-                        match response:
-                            case GenerationResponse():
-                                if shard_metadata.device_rank == 0:
-                                    event_sender.send(
-                                        ChunkGenerated(
-                                            command_id=command_id,
-                                            chunk=TokenChunk(
-                                                idx=response.token,
-                                                model=shard_metadata.model_meta.model_id,
-                                                text=response.text,
-                                                token_id=response.token,
-                                                finish_reason=response.finish_reason,
-                                                stats=response.stats,
-                                            ),
+                    try:
+                        _check_for_debug_prompts(task_params.messages[0].content)
+
+                        # Generate responses using the actual MLX generation
+                        mlx_generator = mlx_generate(
+                            model=model,
+                            tokenizer=tokenizer,
+                            task=task_params,
+                        )
+
+                        # GPT-OSS specific parsing to match other model formats.
+                        if isinstance(model, GptOssModel):
+                            mlx_generator = parse_gpt_oss(mlx_generator)
+
+                        # TODO: Add tool call parser here
+
+                        for response in mlx_generator:
+                            match response:
+                                case GenerationResponse():
+                                    if device_rank == 0:
+                                        event_sender.send(
+                                            ChunkGenerated(
+                                                command_id=command_id,
+                                                chunk=TokenChunk(
+                                                    idx=response.token,
+                                                    model=shard_metadata.model_meta.model_id,
+                                                    text=response.text,
+                                                    token_id=response.token,
+                                                    finish_reason=response.finish_reason,
+                                                    stats=response.stats,
+                                                ),
+                                            )
                                         )
-                                    )
-                                # case TokenizedResponse():
-                                # TODO: something here ig
+
+                    # can we make this more explicit?
+                    except Exception as e:
+                        if device_rank == 0:
+                            event_sender.send(
+                                ChunkGenerated(
+                                    command_id=command_id,
+                                    chunk=TokenChunk(
+                                        idx=0,
+                                        model=shard_metadata.model_meta.model_id,
+                                        text="",
+                                        token_id=0,
+                                        finish_reason="error",
+                                        error_message=str(e),
+                                    ),
+                                )
+                            )
+                        raise
 
                     current_status = RunnerReady()
                     logger.info("runner ready")
@@ -204,7 +248,7 @@ def main(
                 RunnerStatusUpdated(runner_id=runner_id, runner_status=current_status)
             )
             if isinstance(current_status, RunnerShutdown):
-                del model, tokenizer, group, sampler
+                del model, tokenizer, group
                 mx.clear_cache()
                 import gc
 
@@ -212,6 +256,43 @@ def main(
                 break
 
 
+@cache
+def get_gpt_oss_encoding():
+    encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)
+    return encoding
+
+
+def parse_gpt_oss(
+    responses: Generator[GenerationResponse],
+) -> Generator[GenerationResponse]:
+    encoding = get_gpt_oss_encoding()
+    stream = StreamableParser(encoding, role=Role.ASSISTANT)
+    thinking = False
+
+    for response in responses:
+        stream.process(response.token)
+
+        delta = stream.last_content_delta
+        ch = stream.current_channel
+
+        if ch == "analysis" and not thinking:
+            thinking = True
+            yield response.model_copy(update={"text": "<think>"})
+
+        if ch != "analysis" and thinking:
+            thinking = False
+            yield response.model_copy(update={"text": "</think>"})
+
+        if delta:
+            yield response.model_copy(update={"text": delta})
+
+        if response.finish_reason is not None:
+            if thinking:
+                yield response.model_copy(update={"text": "</think>"})
+            yield response
+            break
+
+
 EXO_RUNNER_MUST_FAIL = "EXO RUNNER MUST FAIL"
 EXO_RUNNER_MUST_OOM = "EXO RUNNER MUST OOM"
 EXO_RUNNER_MUST_TIMEOUT = "EXO RUNNER MUST TIMEOUT"

src/exo/worker/tests/unittests/test_mlx/conftest.py

@@ -0,0 +1,201 @@
+# type: ignore
+import json
+import os
+import tempfile
+import traceback
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any, cast
+
+import mlx.core as mx
+import mlx.nn as nn
+
+from exo.shared.constants import EXO_MODELS_DIR
+from exo.shared.types.api import ChatCompletionMessage
+from exo.shared.types.memory import Memory
+from exo.shared.types.models import ModelId, ModelMetadata
+from exo.shared.types.tasks import ChatCompletionTaskParams
+from exo.shared.types.worker.shards import PipelineShardMetadata, TensorShardMetadata
+from exo.worker.engines.mlx import Model
+from exo.worker.engines.mlx.generator.generate import mlx_generate
+from exo.worker.engines.mlx.utils_mlx import shard_and_load
+
+
+class MockLayer(nn.Module):
+    def __init__(self) -> None:
+        super().__init__()
+        self.custom_attr = "test_value"
+        self.use_sliding = True
+
+    def __call__(self, x: mx.array, *args: object, **kwargs: object) -> mx.array:
+        return x * 2
+
+
+@dataclass(frozen=True)
+class PipelineTestConfig:
+    model_path: Path
+    total_layers: int
+    base_port: int
+    max_tokens: int
+
+
+def create_hostfile(world_size: int, base_port: int) -> tuple[str, list[str]]:
+    hosts = [f"127.0.0.1:{base_port + i}" for i in range(world_size)]
+
+    with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as f:
+        json.dump(hosts, f)
+        hostfile_path = f.name
+
+    return hostfile_path, hosts
+
+
+# Use GPT OSS 20b to test as it is a model with a lot of strange behaviour
+
+DEFAULT_GPT_OSS_CONFIG = PipelineTestConfig(
+    model_path=EXO_MODELS_DIR / "mlx-community--gpt-oss-20b-MXFP4-Q8",
+    total_layers=24,
+    base_port=29600,
+    max_tokens=200,
+)
+
+
+DEFAULT_GPT_OSS_MODEL_ID = "mlx-community/gpt-oss-20b-MXFP4-Q8"
+
+
+def run_gpt_oss_pipeline_device(
+    rank: int,
+    world_size: int,
+    hostfile_path: str,
+    layer_splits: list[tuple[int, int]],
+    prompt_tokens: int,
+    prefill_step_size: int,
+    result_queue: Any,  # pyright: ignore[reportAny]
+    max_tokens: int = 200,
+) -> None:
+    os.environ["MLX_HOSTFILE"] = hostfile_path
+    os.environ["MLX_RANK"] = str(rank)
+
+    try:
+        group = mx.distributed.init(backend="ring", strict=True)
+
+        start_layer, end_layer = layer_splits[rank]
+
+        shard_meta = PipelineShardMetadata(
+            model_meta=ModelMetadata(
+                model_id=ModelId(DEFAULT_GPT_OSS_MODEL_ID),
+                pretty_name="GPT-OSS 20B",
+                storage_size=Memory.from_gb(12),
+                n_layers=24,
+                hidden_size=2880,
+                supports_tensor=False,
+            ),
+            device_rank=rank,
+            world_size=world_size,
+            start_layer=start_layer,
+            end_layer=end_layer,
+            n_layers=24,
+        )
+
+        model, tokenizer = shard_and_load(shard_meta, group)
+        model = cast(Model, model)
+
+        # Generate a prompt of exact token length
+        base_text = "The quick brown fox jumps over the lazy dog. "
+        base_tokens = tokenizer.encode(base_text)
+        base_len = len(base_tokens)
+
+        # Build prompt with approximate target length
+        repeats = (prompt_tokens // base_len) + 2
+        long_text = base_text * repeats
+        tokens = tokenizer.encode(long_text)
+        # Truncate to exact target length
+        tokens = tokens[:prompt_tokens]
+        prompt_text = tokenizer.decode(tokens)
+
+        task = ChatCompletionTaskParams(
+            model=DEFAULT_GPT_OSS_MODEL_ID,
+            messages=[ChatCompletionMessage(role="user", content=prompt_text)],
+            max_tokens=max_tokens,
+        )
+
+        generated_text = ""
+        for response in mlx_generate(
+            model=model,
+            tokenizer=tokenizer,
+            task=task,
+        ):
+            generated_text += response.text
+            if response.finish_reason is not None:
+                break
+
+        result_queue.put((rank, True, generated_text))  # pyright: ignore[reportAny]
+
+    except Exception as e:
+        result_queue.put((rank, False, f"{e}\n{traceback.format_exc()}"))  # pyright: ignore[reportAny]
+
+
+def run_gpt_oss_tensor_parallel_device(
+    rank: int,
+    world_size: int,
+    hostfile_path: str,
+    prompt_tokens: int,
+    prefill_step_size: int,
+    result_queue: Any,  # pyright: ignore[reportAny]
+    max_tokens: int = 10,
+) -> None:
+    os.environ["MLX_HOSTFILE"] = hostfile_path
+    os.environ["MLX_RANK"] = str(rank)
+
+    try:
+        group = mx.distributed.init(backend="ring", strict=True)
+
+        # For tensor parallelism, all devices run all layers
+        shard_meta = TensorShardMetadata(
+            model_meta=ModelMetadata(
+                model_id=ModelId(DEFAULT_GPT_OSS_MODEL_ID),
+                pretty_name="GPT-OSS 20B",
+                storage_size=Memory.from_gb(12),
+                n_layers=24,
+                hidden_size=2880,
+                supports_tensor=True,
+            ),
+            device_rank=rank,
+            world_size=world_size,
+            start_layer=0,
+            end_layer=24,
+            n_layers=24,
+        )
+
+        model, tokenizer = shard_and_load(shard_meta, group)
+        model = cast(Model, model)
+
+        base_text = "The quick brown fox jumps over the lazy dog. "
+        base_tokens = tokenizer.encode(base_text)
+        base_len = len(base_tokens)
+
+        repeats = (prompt_tokens // base_len) + 2
+        long_text = base_text * repeats
+        tokens = tokenizer.encode(long_text)
+        tokens = tokens[:prompt_tokens]
+        prompt_text = tokenizer.decode(tokens)
+
+        task = ChatCompletionTaskParams(
+            model=DEFAULT_GPT_OSS_MODEL_ID,
+            messages=[ChatCompletionMessage(role="user", content=prompt_text)],
+            max_tokens=max_tokens,
+        )
+
+        generated_text = ""
+        for response in mlx_generate(
+            model=model,
+            tokenizer=tokenizer,
+            task=task,
+        ):
+            generated_text += response.text
+            if response.finish_reason is not None:
+                break
+
+        result_queue.put((rank, True, generated_text))  # pyright: ignore[reportAny]
+
+    except Exception as e:
+        result_queue.put((rank, False, f"{e}\n{traceback.format_exc()}"))  # pyright: ignore[reportAny]

src/exo/worker/tests/unittests/test_mlx/test_auto_parallel.py

@@ -0,0 +1,146 @@
+import json
+import multiprocessing as mp
+import os
+import tempfile
+from typing import Any
+
+import mlx.core as mx
+import mlx.nn as mlx_nn
+import pytest
+
+from exo.worker.engines.mlx.auto_parallel import (
+    CustomMlxLayer,
+    PipelineFirstLayer,
+    PipelineLastLayer,
+    patch_pipeline_model,
+)
+from exo.worker.tests.unittests.test_mlx.conftest import MockLayer
+
+
+def run_pipeline_device(
+    rank: int,
+    world_size: int,
+    hostfile_path: str,
+    result_queue: Any,  # pyright: ignore[reportAny]
+) -> None:
+    import os
+
+    os.environ["MLX_HOSTFILE"] = hostfile_path
+    os.environ["MLX_RANK"] = str(rank)
+
+    class MockLayerInner(mlx_nn.Module):
+        def __init__(self) -> None:
+            super().__init__()
+            self.custom_attr = "test_value"
+
+        def __call__(self, x: mx.array, *args: object, **kwargs: object) -> mx.array:
+            return x * 2
+
+    class MockModel(mlx_nn.Module):
+        def __init__(self, layers: list[mlx_nn.Module]) -> None:
+            super().__init__()
+            self.layers = layers
+
+        def __call__(self, x: mx.array, *args: object, **kwargs: object) -> mx.array:
+            for layer in self.layers:
+                x = layer(x, *args, **kwargs)  # pyright: ignore[reportUnknownVariableType]
+            return x  # pyright: ignore[reportUnknownVariableType]
+
+    try:
+        group = mx.distributed.init(backend="ring", strict=True)
+
+        mock = MockLayerInner()
+        first = PipelineFirstLayer(mock, r=rank, group=group)
+        composed = PipelineLastLayer(first, r=rank, s=world_size, group=group)
+
+        # Wrap in a mock model, then wrap in PipelineParallelModel for all_gather
+        inner_model = MockModel([composed])
+        model = patch_pipeline_model(inner_model, group)
+
+        x = mx.ones((1, 4))
+        result = model(x)
+        mx.eval(result)
+        success = result.shape == x.shape
+        result_queue.put((rank, success, result))  # pyright: ignore[reportAny]
+    except Exception as e:
+        result_queue.put((rank, False, str(e)))  # pyright: ignore[reportAny]
+
+
+def test_single_wrapper_delegates_attributes() -> None:
+    mock = MockLayer()
+    wrapped = CustomMlxLayer(mock)
+
+    assert wrapped.custom_attr == "test_value"  # type: ignore[attr-defined]
+    assert wrapped.use_sliding is True  # type: ignore[attr-defined]
+
+
+def test_composed_wrappers_delegate_attributes() -> None:
+    mock = MockLayer()
+    group = mx.distributed.init()
+
+    first = PipelineFirstLayer(mock, r=0, group=group)
+    composed = PipelineLastLayer(first, r=0, s=1, group=group)
+
+    assert composed.custom_attr == "test_value"  # type: ignore[attr-defined]
+    assert composed.use_sliding is True  # type: ignore[attr-defined]
+
+
+def test_missing_attribute_raises() -> None:
+    mock = MockLayer()
+    wrapped = CustomMlxLayer(mock)
+
+    with pytest.raises(AttributeError):
+        _ = wrapped.nonexistent_attr  # type: ignore[attr-defined]
+
+
+def test_composed_call_works() -> None:
+    ctx = mp.get_context("spawn")
+
+    world_size = 2
+    base_port = 29500
+
+    hosts = [f"127.0.0.1:{base_port + i}" for i in range(world_size)]
+
+    with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as f:
+        json.dump(hosts, f)
+        hostfile_path = f.name
+
+    try:
+        result_queue: Any = ctx.Queue()
+
+        processes: list[Any] = []
+        for rank in range(world_size):
+            p = ctx.Process(
+                target=run_pipeline_device,
+                args=(rank, world_size, hostfile_path, result_queue),
+            )
+            p.start()
+            processes.append(p)
+
+        for p in processes:  # pyright: ignore[reportAny]
+            p.join(timeout=10)  # pyright: ignore[reportAny]
+
+        results: dict[int, Any] = {}
+        errors: dict[int, str] = {}
+        while not result_queue.empty():  # pyright: ignore[reportAny]
+            rank, success, value = result_queue.get()  # pyright: ignore[reportAny]
+            if success:
+                results[rank] = value
+            else:
+                errors[rank] = value
+
+        assert len(results) == world_size, (
+            f"Expected {world_size} results, got {len(results)}. Errors: {errors}"
+        )
+
+        for rank in range(world_size):
+            assert rank in results, (
+                f"Device {rank} failed: {errors.get(rank, 'unknown')}"
+            )
+            result_array = results[rank]
+            # Both devices see the final result (4.0) after all_gather
+            assert (result_array == 4.0).all(), (
+                f"Device {rank}: expected 4.0, got {result_array}"
+            )
+    finally:
+        os.unlink(hostfile_path)

src/exo/worker/tests/unittests/test_mlx/test_distributed_fix.py

@@ -0,0 +1,230 @@
+import multiprocessing as mp
+import os
+from dataclasses import dataclass
+from typing import Any, Callable
+
+import pytest
+
+from exo.worker.tests.unittests.test_mlx.conftest import (
+    DEFAULT_GPT_OSS_CONFIG,
+    create_hostfile,
+    run_gpt_oss_pipeline_device,
+    run_gpt_oss_tensor_parallel_device,
+)
+
+
+def _check_model_exists() -> bool:
+    return DEFAULT_GPT_OSS_CONFIG.model_path.exists()
+
+
+pytestmark = [
+    pytest.mark.skipif(
+        not _check_model_exists(),
+        reason=f"GPT-OSS model not found at {DEFAULT_GPT_OSS_CONFIG.model_path}",
+    ),
+]
+
+
+@dataclass
+class DistributedTestResult:
+    timed_out: bool
+    world_size: int
+    results: dict[int, tuple[bool, str]]
+
+    @property
+    def all_success(self) -> bool:
+        if len(self.results) != self.world_size:
+            return False
+        return all(r[0] for r in self.results.values())
+
+
+def run_distributed_test(
+    world_size: int,
+    port_offset: int,
+    process_timeout: int,
+    target: Callable[..., None],
+    make_args: Callable[[int], tuple[Any, ...]],
+) -> DistributedTestResult:
+    ctx = mp.get_context("spawn")
+    hostfile_path, _ = create_hostfile(
+        world_size, DEFAULT_GPT_OSS_CONFIG.base_port + port_offset
+    )
+
+    try:
+        result_queue: Any = ctx.Queue()
+        processes: list[Any] = []
+
+        for rank in range(world_size):
+            args = make_args(rank)
+            p = ctx.Process(
+                target=target,
+                args=(rank, world_size, hostfile_path, *args, result_queue),
+            )
+            p.start()
+            processes.append(p)
+
+        for p in processes:  # pyright: ignore[reportAny]
+            p.join(timeout=process_timeout)  # pyright: ignore[reportAny]
+
+        timed_out = any(p.is_alive() for p in processes)  # pyright: ignore[reportAny]
+
+        for p in processes:  # pyright: ignore[reportAny]
+            if p.is_alive():  # pyright: ignore[reportAny]
+                p.terminate()  # pyright: ignore[reportAny]
+                p.join(timeout=5)  # pyright: ignore[reportAny]
+
+        results: dict[int, tuple[bool, str]] = {}
+        while not result_queue.empty():  # pyright: ignore[reportAny]
+            rank, success, value = result_queue.get()  # pyright: ignore[reportAny]
+            results[rank] = (success, value)
+
+        return DistributedTestResult(
+            timed_out=timed_out, world_size=world_size, results=results
+        )
+
+    finally:
+        os.unlink(hostfile_path)
+
+
+def run_pipeline_test(
+    layer_splits: list[tuple[int, int]],
+    prompt_tokens: int,
+    prefill_step_size: int,
+    port_offset: int = 0,
+    process_timeout: int = 60,
+) -> DistributedTestResult:
+    def make_args(rank: int) -> tuple[Any, ...]:
+        return (
+            layer_splits,
+            prompt_tokens,
+            prefill_step_size,
+        )
+
+    return run_distributed_test(
+        world_size=len(layer_splits),
+        port_offset=port_offset,
+        process_timeout=process_timeout,
+        target=run_gpt_oss_pipeline_device,
+        make_args=make_args,
+    )
+
+
+def run_tensor_test(
+    prompt_tokens: int,
+    prefill_step_size: int,
+    port_offset: int = 0,
+    process_timeout: int = 60,
+) -> DistributedTestResult:
+    def make_args(rank: int) -> tuple[Any, ...]:
+        return (
+            prompt_tokens,
+            prefill_step_size,
+        )
+
+    return run_distributed_test(
+        world_size=2,
+        port_offset=port_offset,
+        process_timeout=process_timeout,
+        target=run_gpt_oss_tensor_parallel_device,
+        make_args=make_args,
+    )
+
+
+class TestPipelineParallelFix:
+    BUG_TRIGGER_SPLITS: list[tuple[int, int]] = [(0, 1), (1, 24)]
+
+    def test_pipeline_single_layer_first_device(self) -> None:
+        result = run_pipeline_test(
+            layer_splits=self.BUG_TRIGGER_SPLITS,
+            prompt_tokens=100,
+            prefill_step_size=64,
+            process_timeout=60,
+        )
+        assert not result.timed_out, "Unexpected timeout - fix may not be working"
+        assert result.all_success, f"Failures: {result.results}"
+
+
+class TestPipelineSplitConfigurations:
+    @pytest.mark.parametrize(
+        "layer_splits",
+        [
+            [(0, 1), (1, 24)],
+            [(0, 6), (6, 24)],
+            [(0, 12), (12, 24)],
+        ],
+        ids=["1_23", "6_18", "12_12"],
+    )
+    def test_pipeline_splits(
+        self,
+        layer_splits: list[tuple[int, int]],
+    ) -> None:
+        result = run_pipeline_test(
+            layer_splits=layer_splits,
+            prompt_tokens=600,
+            prefill_step_size=512,
+            port_offset=100,
+        )
+        assert not result.timed_out, f"Timeout with {layer_splits}"
+        assert result.all_success, f"Failures with {layer_splits}: {result.results}"
+
+
+class TestPrefillStepSizeBoundaries:
+    @pytest.mark.parametrize(
+        "prefill_step_size,prompt_tokens",
+        [
+            (512, 511),
+            (512, 512),
+            (512, 513),
+            (512, 1024),
+        ],
+        ids=["under", "exact", "over", "double"],
+    )
+    def test_boundary_conditions(
+        self,
+        prefill_step_size: int,
+        prompt_tokens: int,
+    ) -> None:
+        result = run_pipeline_test(
+            layer_splits=[(0, 12), (12, 24)],
+            prompt_tokens=prompt_tokens,
+            prefill_step_size=prefill_step_size,
+            port_offset=200,
+        )
+        assert not result.timed_out, f"Timeout: {prompt_tokens=}, {prefill_step_size=}"
+        assert result.all_success, f"Failures: {result.results}"
+
+
+class TestTensorParallelFix:
+    def test_tensor_parallel(self) -> None:
+        result = run_tensor_test(
+            prompt_tokens=100,
+            prefill_step_size=64,
+            port_offset=400,
+        )
+        assert not result.timed_out, "Unexpected timeout"
+        assert result.all_success, f"Failures: {result.results}"
+
+
+class TestTensorParallelBoundaries:
+    @pytest.mark.parametrize(
+        "prefill_step_size,prompt_tokens",
+        [
+            (512, 511),
+            (512, 512),
+            (512, 513),
+            (512, 1024),
+        ],
+        ids=["under", "exact", "over", "double"],
+    )
+    def test_tensor_parallel_boundaries(
+        self,
+        prefill_step_size: int,
+        prompt_tokens: int,
+    ) -> None:
+        result = run_tensor_test(
+            prompt_tokens=prompt_tokens,
+            prefill_step_size=prefill_step_size,
+            port_offset=500,
+        )
+        assert not result.timed_out, f"Timeout: {prompt_tokens=}, {prefill_step_size=}"
+        assert result.all_success, f"Failures: {result.results}"

src/exo/worker/tests/unittests/test_mlx/test_tokenizers.py

@@ -0,0 +1,386 @@
+"""
+Unit tests for tokenizer loading and functionality across all supported models.
+
+This test downloads only tokenizer-related files (not full model weights) to verify
+that tokenizers can be loaded and used correctly for encoding/decoding.
+"""
+
+import asyncio
+import contextlib
+from pathlib import Path
+
+import pytest
+
+from exo.shared.models.model_cards import MODEL_CARDS, ModelCard
+from exo.worker.download.download_utils import (
+    download_file_with_retry,
+    ensure_models_dir,
+    fetch_file_list_with_cache,
+)
+from exo.worker.engines.mlx.utils_mlx import (
+    get_eos_token_ids_for_model,
+    load_tokenizer_for_model_id,
+)
+
+# Files needed for tokenizer functionality
+TOKENIZER_FILE_PATTERNS = [
+    "tokenizer.json",
+    "tokenizer_config.json",
+    "special_tokens_map.json",
+    "vocab.json",
+    "vocab.txt",
+    "merges.txt",
+    "tiktoken.model",
+    "added_tokens.json",
+    "tokenizer.model",
+    "tokenization_*.py",  # Custom tokenizer implementations
+]
+
+
+def is_tokenizer_file(filename: str) -> bool:
+    """Check if a file is needed for tokenizer functionality."""
+    for pattern in TOKENIZER_FILE_PATTERNS:
+        if "*" in pattern:
+            prefix = pattern.split("*")[0]
+            suffix = pattern.split("*")[1]
+            if filename.startswith(prefix) and filename.endswith(suffix):
+                return True
+        elif filename == pattern:
+            return True
+    return False
+
+
+async def download_tokenizer_files(model_id: str) -> Path:
+    """Download only the tokenizer-related files for a model."""
+    target_dir = await ensure_models_dir() / model_id.replace("/", "--")
+    target_dir.mkdir(parents=True, exist_ok=True)
+
+    file_list = await fetch_file_list_with_cache(model_id, "main", recursive=True)
+
+    tokenizer_files = [f for f in file_list if is_tokenizer_file(f.path)]
+
+    if not tokenizer_files:
+        pytest.skip(f"No tokenizer files found for {model_id}")
+
+    for file_entry in tokenizer_files:
+        with contextlib.suppress(FileNotFoundError):
+            await download_file_with_retry(
+                model_id, "main", file_entry.path, target_dir
+            )
+
+    return target_dir
+
+
+# Get a sample of models to test (one per family to keep tests fast)
+def get_test_models() -> list[tuple[str, ModelCard]]:
+    """Get a representative sample of models to test."""
+    # Pick one model from each family to test
+    families: dict[str, tuple[str, ModelCard]] = {}
+    for short_id, card in MODEL_CARDS.items():
+        # Extract family name (e.g., "llama-3.1" from "llama-3.1-8b")
+        parts = short_id.split("-")
+        family = "-".join(parts[:2]) if len(parts) >= 2 else parts[0]
+
+        if family not in families:
+            families[family] = (short_id, card)
+
+    return list(families.values())
+
+
+TEST_MODELS: list[tuple[str, ModelCard]] = get_test_models()
+
+
+@pytest.fixture(scope="module")
+def event_loop():
+    """Create event loop for async tests."""
+    loop = asyncio.new_event_loop()
+    yield loop
+    loop.close()
+
+
+@pytest.mark.parametrize(
+    "short_id,model_card",
+    TEST_MODELS,
+    ids=[m[0] for m in TEST_MODELS],
+)
+@pytest.mark.asyncio
+async def test_tokenizer_encode_decode(short_id: str, model_card: ModelCard) -> None:
+    """Test that tokenizer can encode and decode text correctly."""
+    model_id = str(model_card.model_id)
+
+    # Download tokenizer files
+    model_path = await download_tokenizer_files(model_id)
+
+    # Verify required files exist
+    has_tokenizer = (
+        (model_path / "tokenizer.json").exists()
+        or (model_path / "tokenizer_config.json").exists()
+        or (model_path / "tiktoken.model").exists()
+        or (model_path / "tokenizer.model").exists()
+    )
+    if not has_tokenizer:
+        pytest.skip(f"Required tokenizer files not found for {model_id}")
+
+    # Load tokenizer
+    tokenizer = load_tokenizer_for_model_id(model_id, model_path)
+
+    # Test basic encoding
+    test_text = "Hello, world!"
+    encoded = tokenizer.encode(test_text)
+    assert isinstance(encoded, list), f"encode() should return a list for {model_id}"
+    assert len(encoded) > 0, f"encode() should return non-empty list for {model_id}"
+    assert all(isinstance(t, int) for t in encoded), (
+        f"All tokens should be integers for {model_id}"
+    )
+
+    # Test decoding
+    decoded = tokenizer.decode(encoded)
+    assert isinstance(decoded, str), f"decode() should return a string for {model_id}"
+    assert test_text in decoded or decoded.strip() == test_text.strip(), (
+        f"decode(encode(x)) should preserve text for {model_id}: got {decoded!r}"
+    )
+
+    # Test with longer text
+    long_text = "The quick brown fox jumps over the lazy dog. " * 10
+    long_encoded = tokenizer.encode(long_text)
+    assert len(long_encoded) > len(encoded), (
+        f"Longer text should produce more tokens for {model_id}"
+    )
+
+    # Test empty string
+    empty_encoded = tokenizer.encode("")
+    assert isinstance(empty_encoded, list), (
+        f"encode('') should return a list for {model_id}"
+    )
+
+    # Test special characters
+    special_text = 'Hello!\n\tWorld? <test> & "quotes"'
+    special_encoded = tokenizer.encode(special_text)
+    assert len(special_encoded) > 0, f"Special chars should encode for {model_id}"
+
+    # Test unicode
+    unicode_text = "Hello 世界 🌍"
+    unicode_encoded = tokenizer.encode(unicode_text)
+    assert len(unicode_encoded) > 0, f"Unicode should encode for {model_id}"
+
+
+@pytest.mark.parametrize(
+    "short_id,model_card",
+    TEST_MODELS,
+    ids=[m[0] for m in TEST_MODELS],
+)
+@pytest.mark.asyncio
+async def test_tokenizer_has_required_attributes(
+    short_id: str, model_card: ModelCard
+) -> None:
+    """Test that tokenizer has required attributes for inference."""
+    model_id = str(model_card.model_id)
+
+    model_path = await download_tokenizer_files(model_id)
+
+    has_tokenizer = (
+        (model_path / "tokenizer.json").exists()
+        or (model_path / "tokenizer_config.json").exists()
+        or (model_path / "tiktoken.model").exists()
+        or (model_path / "tokenizer.model").exists()
+    )
+    if not has_tokenizer:
+        pytest.skip(f"Required tokenizer files not found for {model_id}")
+
+    tokenizer = load_tokenizer_for_model_id(model_id, model_path)
+    eos_token_ids = get_eos_token_ids_for_model(model_id)
+
+    # Check for vocabulary size
+    empty_vocab: dict[str, int] = {}
+    vocab_size: int = getattr(tokenizer, "vocab_size", None) or len(
+        getattr(tokenizer, "get_vocab", lambda: empty_vocab)()
+    )
+    assert vocab_size > 0, f"Tokenizer should have vocab_size > 0 for {model_id}"
+
+    # Check for EOS token (either from tokenizer or explicitly provided)
+    has_eos = (
+        eos_token_ids is not None
+        or getattr(tokenizer, "eos_token_id", None) is not None
+        or getattr(tokenizer, "eos_token", None) is not None
+    )
+    assert has_eos, f"Tokenizer should have EOS token for {model_id}"
+
+
+@pytest.mark.parametrize(
+    "short_id,model_card",
+    TEST_MODELS,
+    ids=[m[0] for m in TEST_MODELS],
+)
+@pytest.mark.asyncio
+async def test_tokenizer_special_tokens(short_id: str, model_card: ModelCard) -> None:
+    """Test that tokenizer can encode text containing special tokens.
+
+    This is critical because the actual inference path uses prompts with
+    special tokens from chat templates. If special tokens aren't handled
+    correctly, encoding will fail.
+    """
+    model_id = str(model_card.model_id)
+
+    model_path = await download_tokenizer_files(model_id)
+
+    has_tokenizer = (
+        (model_path / "tokenizer.json").exists()
+        or (model_path / "tokenizer_config.json").exists()
+        or (model_path / "tiktoken.model").exists()
+        or (model_path / "tokenizer.model").exists()
+    )
+    assert has_tokenizer, f"Required tokenizer files not found for {model_id}"
+
+    tokenizer = load_tokenizer_for_model_id(model_id, model_path)
+
+    # Get special tokens from the tokenizer
+    special_tokens: list[str] = []
+
+    # Try to get special tokens from various sources
+    if hasattr(tokenizer, "all_special_tokens"):
+        special_tokens.extend(tokenizer.all_special_tokens)
+    elif hasattr(tokenizer, "_tokenizer") and hasattr(
+        tokenizer._tokenizer,
+        "all_special_tokens",
+    ):
+        special_tokens.extend(tokenizer._tokenizer.all_special_tokens)
+
+    # Also check for common special token attributes
+    for attr in [
+        "bos_token",
+        "eos_token",
+        "pad_token",
+        "unk_token",
+        "sep_token",
+        "cls_token",
+    ]:
+        token = getattr(tokenizer, attr, None)
+        if token is None and hasattr(tokenizer, "_tokenizer"):
+            token = getattr(tokenizer._tokenizer, attr, None)
+        if token and isinstance(token, str) and token not in special_tokens:
+            special_tokens.append(token)
+
+    # If we found special tokens, test encoding text that contains them
+    if special_tokens:
+        # Create text with special tokens interspersed
+        test_with_special = f"{special_tokens[0]}Hello world"
+        if len(special_tokens) > 1:
+            test_with_special += f"{special_tokens[1]}"
+
+        encoded = tokenizer.encode(test_with_special)
+        assert isinstance(encoded, list), (
+            f"encode() with special tokens should return list for {model_id}"
+        )
+        assert len(encoded) > 0, (
+            f"encode() with special tokens should return non-empty list for {model_id}"
+        )
+        assert all(isinstance(t, int) for t in encoded), (
+            f"All tokens should be integers for {model_id}"
+        )
+
+        # Verify we can decode
+        decoded = tokenizer.decode(encoded)
+        assert isinstance(decoded, str), f"decode() should return string for {model_id}"
+
+    # Test with angle-bracket tokens (common format for special tokens)
+    # These should not raise errors even if they're not actual special tokens
+    angle_bracket_text = "<|test|>Hello<|end|>"
+    encoded = tokenizer.encode(angle_bracket_text)
+    assert isinstance(encoded, list), (
+        f"encode() with angle brackets should return list for {model_id}"
+    )
+    assert len(encoded) > 0, (
+        f"encode() with angle brackets should be non-empty for {model_id}"
+    )
+
+
+# Specifically test Kimi tokenizer since it has special handling
+@pytest.mark.asyncio
+async def test_kimi_tokenizer_specifically():
+    """Test Kimi tokenizer with its specific patches and quirks."""
+    kimi_models = [
+        (short_id, card)
+        for short_id, card in MODEL_CARDS.items()
+        if "kimi" in short_id.lower()
+    ]
+
+    if not kimi_models:
+        pytest.skip("No Kimi models found in MODEL_CARDS")
+
+    _, model_card = kimi_models[0]
+    model_id = str(model_card.model_id)
+
+    model_path = await download_tokenizer_files(model_id)
+
+    # Ensure the custom tokenizer file exists
+    if not (model_path / "tokenization_kimi.py").exists():
+        pytest.skip("tokenization_kimi.py not found")
+
+    tokenizer = load_tokenizer_for_model_id(model_id, model_path)
+    eos_token_ids = get_eos_token_ids_for_model(model_id)
+
+    # Test encode/decode cycle
+    test_text = "Hello, world!"
+    encoded = tokenizer.encode(test_text)
+    decoded = tokenizer.decode(encoded)
+
+    assert len(encoded) > 0, "Kimi tokenizer should encode text"
+    assert isinstance(decoded, str), "Kimi tokenizer should decode to string"
+
+    # Test that the patched encode works (returns list of ints)
+    assert all(isinstance(t, int) for t in encoded), "Tokens should be integers"
+
+    # Test encoding text with special tokens (like from chat templates)
+    # This is critical - the warmup inference uses prompts with special tokens
+    special_token_text = "<|im_user|>user<|im_middle|>Hello<|im_end|><|im_assistant|>"
+    special_encoded = tokenizer.encode(special_token_text)
+    assert len(special_encoded) > 0, "Kimi tokenizer should handle special tokens"
+    assert all(isinstance(t, int) for t in special_encoded), (
+        "Special token encoding should return integers"
+    )
+
+    # Verify EOS token is set
+    assert eos_token_ids == [163586], "Kimi EOS token should be [163586]"
+
+
+# Test GLM tokenizer since it also has special handling
+@pytest.mark.asyncio
+async def test_glm_tokenizer_specifically():
+    """Test GLM tokenizer with its specific EOS tokens."""
+    glm_models = [
+        (short_id, card)
+        for short_id, card in MODEL_CARDS.items()
+        if "glm" in short_id.lower()
+    ]
+
+    if not glm_models:
+        pytest.skip("No GLM models found in MODEL_CARDS")
+
+    _, model_card = glm_models[0]
+    model_id = str(model_card.model_id)
+
+    model_path = await download_tokenizer_files(model_id)
+
+    has_tokenizer = (model_path / "tokenizer.json").exists() or (
+        model_path / "tokenizer_config.json"
+    ).exists()
+    if not has_tokenizer:
+        pytest.skip("GLM tokenizer files not found")
+
+    tokenizer = load_tokenizer_for_model_id(model_id, model_path)
+    eos_token_ids = get_eos_token_ids_for_model(model_id)
+
+    # Test encode/decode
+    test_text = "Hello, world!"
+    encoded = tokenizer.encode(test_text)
+    decoded = tokenizer.decode(encoded)
+
+    assert len(encoded) > 0, "GLM tokenizer should encode text"
+    assert isinstance(decoded, str), "GLM tokenizer should decode to string"
+
+    # Verify EOS tokens
+    assert eos_token_ids == [
+        151336,
+        151329,
+        151338,
+    ], "GLM EOS tokens should be correct"

src/exo/worker/tests/unittests/test_plan/test_download_and_loading.py

@@ -1,5 +1,6 @@
 import exo.worker.plan as plan_mod
 from exo.shared.types.common import NodeId
+from exo.shared.types.memory import Memory
 from exo.shared.types.models import ModelId
 from exo.shared.types.tasks import LoadModel
 from exo.shared.types.worker.downloads import DownloadCompleted, DownloadProgress
@@ -94,13 +95,23 @@ def test_plan_loads_model_when_all_shards_downloaded_and_waiting():
 
     # Local node has already marked its shard as downloaded (not actually used by _load_model)
     local_download_status = {
-        MODEL_A_ID: DownloadCompleted(shard_metadata=shard1, node_id=NODE_A)
+        MODEL_A_ID: DownloadCompleted(
+            shard_metadata=shard1, node_id=NODE_A, total_bytes=Memory()
+        )
     }
 
     # Global view has completed downloads for both nodes
     global_download_status = {
-        NODE_A: [DownloadCompleted(shard_metadata=shard1, node_id=NODE_A)],
-        NODE_B: [DownloadCompleted(shard_metadata=shard2, node_id=NODE_B)],
+        NODE_A: [
+            DownloadCompleted(
+                shard_metadata=shard1, node_id=NODE_A, total_bytes=Memory()
+            )
+        ],
+        NODE_B: [
+            DownloadCompleted(
+                shard_metadata=shard2, node_id=NODE_B, total_bytes=Memory()
+            )
+        ],
     }
 
     result = plan_mod.plan(
@@ -140,7 +151,9 @@ def test_plan_does_not_request_download_when_shard_already_downloaded():
 
     # Local status claims the shard is downloaded already
     local_download_status = {
-        MODEL_A_ID: DownloadCompleted(shard_metadata=shard, node_id=NODE_A)
+        MODEL_A_ID: DownloadCompleted(
+            shard_metadata=shard, node_id=NODE_A, total_bytes=Memory()
+        )
     }
 
     # Global view hasn't caught up yet (no completed shards recorded for NODE_A)
@@ -192,10 +205,16 @@ def test_plan_does_not_load_model_until_all_shards_downloaded_globally():
 
     # Only NODE_A's shard is recorded as downloaded globally
     local_download_status = {
-        MODEL_A_ID: DownloadCompleted(shard_metadata=shard1, node_id=NODE_A)
+        MODEL_A_ID: DownloadCompleted(
+            shard_metadata=shard1, node_id=NODE_A, total_bytes=Memory()
+        )
     }
     global_download_status = {
-        NODE_A: [DownloadCompleted(shard_metadata=shard1, node_id=NODE_A)],
+        NODE_A: [
+            DownloadCompleted(
+                shard_metadata=shard1, node_id=NODE_A, total_bytes=Memory()
+            )
+        ],
         NODE_B: [],  # NODE_B has no downloads completed yet
     }
 
@@ -212,9 +231,15 @@ def test_plan_does_not_load_model_until_all_shards_downloaded_globally():
     assert result is None
 
     global_download_status = {
-        NODE_A: [DownloadCompleted(shard_metadata=shard1, node_id=NODE_A)],
+        NODE_A: [
+            DownloadCompleted(
+                shard_metadata=shard1, node_id=NODE_A, total_bytes=Memory()
+            )
+        ],
         NODE_B: [
-            DownloadCompleted(shard_metadata=shard2, node_id=NODE_B)
+            DownloadCompleted(
+                shard_metadata=shard2, node_id=NODE_B, total_bytes=Memory()
+            )
         ],  # NODE_B has no downloads completed yet
     }
 

src/exo/worker/tests/unittests/test_runner/test_event_ordering.py

@@ -111,7 +111,7 @@ def assert_events_equal(test_events: Iterable[Event], true_events: Iterable[Even
 def patch_out_mlx(monkeypatch: pytest.MonkeyPatch):
     # initialize_mlx returns a "group" equal to 1
     monkeypatch.setattr(mlx_runner, "initialize_mlx", make_nothin(1))
-    monkeypatch.setattr(mlx_runner, "load_mlx_items", make_nothin((1, 1, 1)))
+    monkeypatch.setattr(mlx_runner, "load_mlx_items", make_nothin((1, 1)))
     monkeypatch.setattr(mlx_runner, "warmup_inference", make_nothin(1))
     monkeypatch.setattr(mlx_runner, "_check_for_debug_prompts", nothin)
 
@@ -121,6 +121,21 @@ def patch_out_mlx(monkeypatch: pytest.MonkeyPatch):
     monkeypatch.setattr(mlx_runner, "mlx_generate", fake_generate)
 
 
+# Use a fake event_sender to remove test flakiness.
+class EventCollector:
+    def __init__(self) -> None:
+        self.events: list[Event] = []
+
+    def send(self, event: Event) -> None:
+        self.events.append(event)
+
+    def close(self) -> None:
+        pass
+
+    def join(self) -> None:
+        pass
+
+
 def _run(tasks: Iterable[Task]):
     bound_instance = get_bound_mlx_ring_instance(
         instance_id=INSTANCE_1_ID,
@@ -130,22 +145,20 @@ def _run(tasks: Iterable[Task]):
     )
 
     task_sender, task_receiver = mp_channel[Task]()
-    event_sender, event_receiver = mp_channel[Event]()
+    event_sender = EventCollector()
 
-    with task_sender, event_receiver:
+    with task_sender:
         for t in tasks:
             task_sender.send(t)
 
         # worst monkeypatch known to man
         # this is some c++ nonsense
-        event_sender.close = nothin
-        event_sender.join = nothin
         task_receiver.close = nothin
         task_receiver.join = nothin
 
-        mlx_runner.main(bound_instance, event_sender, task_receiver)
+        mlx_runner.main(bound_instance, event_sender, task_receiver)  # type: ignore[arg-type]
 
-        return event_receiver.collect()
+        return event_sender.events
 
 
 def test_events_processed_in_correct_order(patch_out_mlx: pytest.MonkeyPatch):

src/exo/worker/utils/__init__.py

@@ -1,6 +0,0 @@
-from .profile import start_polling_memory_metrics, start_polling_node_metrics
-
-__all__ = [
-    "start_polling_node_metrics",
-    "start_polling_memory_metrics",
-]

src/exo/worker/utils/macmon.py

@@ -1,103 +0,0 @@
-import platform
-import shutil
-from subprocess import CalledProcessError
-from typing import cast
-
-from anyio import run_process
-from pydantic import BaseModel, ConfigDict, ValidationError
-
-
-class MacMonError(Exception):
-    """Exception raised for errors in the MacMon functions."""
-
-
-def _get_binary_path() -> str:
-    """
-    Get the path to the macmon binary.
-
-    Raises:
-        MacMonError: If the binary doesn't exist or can't be made executable.
-    """
-    # Check for macOS with ARM chip
-    system = platform.system().lower()
-    machine = platform.machine().lower()
-
-    if system != "darwin" or not (
-        "arm" in machine or "m1" in machine or "m2" in machine
-    ):
-        raise MacMonError("MacMon only supports macOS with Apple Silicon (ARM) chips")
-
-    path = shutil.which("macmon")
-
-    if path is None:
-        raise MacMonError("MacMon not found in PATH")
-
-    return path
-
-
-class TempMetrics(BaseModel):
-    """Temperature-related metrics returned by macmon."""
-
-    cpu_temp_avg: float
-    gpu_temp_avg: float
-
-    model_config = ConfigDict(extra="ignore")
-
-
-class Metrics(BaseModel):
-    """Complete set of metrics returned by macmon.
-
-    Unknown fields are ignored for forward-compatibility.
-    """
-
-    all_power: float
-    ane_power: float
-    cpu_power: float
-    ecpu_usage: tuple[int, float]
-    gpu_power: float
-    gpu_ram_power: float
-    gpu_usage: tuple[int, float]
-    pcpu_usage: tuple[int, float]
-    ram_power: float
-    sys_power: float
-    temp: TempMetrics
-    timestamp: str
-
-    model_config = ConfigDict(extra="ignore")
-
-
-async def get_metrics_async() -> Metrics:
-    """
-    Asynchronously run the binary and return the metrics as a Python dictionary.
-
-    Args:
-        binary_path: Optional path to the binary. If not provided, will use the bundled binary.
-
-    Returns:
-        A mapping containing system metrics.
-
-    Raises:
-        MacMonError: If there's an error running the binary.
-    """
-    path = _get_binary_path()
-
-    try:
-        # TODO: Keep Macmon running in the background?
-        result = await run_process([path, "pipe", "-s", "1"])
-
-        return Metrics.model_validate_json(result.stdout.decode().strip())
-
-    except ValidationError as e:
-        raise MacMonError(f"Error parsing JSON output: {e}") from e
-    except CalledProcessError as e:
-        stderr_msg = "no stderr"
-        stderr_output = cast(bytes | str | None, e.stderr)
-        if stderr_output is not None:
-            stderr_msg = (
-                stderr_output.decode()
-                if isinstance(stderr_output, bytes)
-                else str(stderr_output)
-            )
-        raise MacMonError(
-            f"MacMon failed with return code {e.returncode}: {stderr_msg}"
-        ) from e

src/exo/worker/utils/net_profile.py

@@ -1,78 +0,0 @@
-import http.client
-
-from anyio import create_task_group, to_thread
-from loguru import logger
-
-from exo.shared.topology import Topology
-from exo.shared.types.common import NodeId
-
-
-async def check_reachability(
-    target_ip: str,
-    expected_node_id: NodeId,
-    self_node_id: NodeId,
-    out: dict[NodeId, set[str]],
-) -> None:
-    """Check if a node is reachable at the given IP and verify its identity."""
-
-    def _fetch_remote_node_id() -> NodeId | None:
-        connection = http.client.HTTPConnection(target_ip, 52415, timeout=1)
-        try:
-            connection.request("GET", "/node_id")
-            response = connection.getresponse()
-            if response.status != 200:
-                return None
-
-            body = response.read().decode("utf-8").strip()
-
-            # Strip quotes if present (JSON string response)
-            if body.startswith('"') and body.endswith('"') and len(body) >= 2:
-                body = body[1:-1]
-
-            return NodeId(body) or None
-        except OSError:
-            return None
-        except http.client.HTTPException:
-            return None
-        finally:
-            connection.close()
-
-    remote_node_id = await to_thread.run_sync(_fetch_remote_node_id)
-    if remote_node_id is None:
-        return
-
-    if remote_node_id == self_node_id:
-        return
-
-    if remote_node_id != expected_node_id:
-        logger.warning(
-            f"Discovered node with unexpected node_id; "
-            f"ip={target_ip}, expected_node_id={expected_node_id}, "
-            f"remote_node_id={remote_node_id}"
-        )
-        return
-
-    if remote_node_id not in out:
-        out[remote_node_id] = set()
-    out[remote_node_id].add(target_ip)
-
-
-async def check_reachable(
-    topology: Topology, self_node_id: NodeId
-) -> dict[NodeId, set[str]]:
-    """Check which nodes are reachable and return their IPs."""
-    reachable: dict[NodeId, set[str]] = {}
-    async with create_task_group() as tg:
-        for node in topology.list_nodes():
-            if not node.node_profile:
-                continue
-            for iface in node.node_profile.network_interfaces:
-                tg.start_soon(
-                    check_reachability,
-                    iface.ip_address,
-                    node.node_id,
-                    self_node_id,
-                    reachable,
-                )
-
-    return reachable

src/exo/worker/utils/profile.py

@@ -1,114 +0,0 @@
-import asyncio
-import os
-import platform
-from typing import Any, Callable, Coroutine
-
-import anyio
-from loguru import logger
-
-from exo.shared.types.memory import Memory
-from exo.shared.types.profiling import (
-    MemoryPerformanceProfile,
-    NodePerformanceProfile,
-    SystemPerformanceProfile,
-)
-
-from .macmon import (
-    MacMonError,
-    Metrics,
-)
-from .macmon import (
-    get_metrics_async as macmon_get_metrics_async,
-)
-from .system_info import (
-    get_friendly_name,
-    get_model_and_chip,
-    get_network_interfaces,
-)
-
-
-async def get_metrics_async() -> Metrics | None:
-    """Return detailed Metrics on macOS or a minimal fallback elsewhere."""
-
-    if platform.system().lower() == "darwin":
-        return await macmon_get_metrics_async()
-
-
-def get_memory_profile() -> MemoryPerformanceProfile:
-    """Construct a MemoryPerformanceProfile using psutil"""
-    override_memory_env = os.getenv("OVERRIDE_MEMORY_MB")
-    override_memory: int | None = (
-        Memory.from_mb(int(override_memory_env)).in_bytes
-        if override_memory_env
-        else None
-    )
-
-    return MemoryPerformanceProfile.from_psutil(override_memory=override_memory)
-
-
-async def start_polling_memory_metrics(
-    callback: Callable[[MemoryPerformanceProfile], Coroutine[Any, Any, None]],
-    *,
-    poll_interval_s: float = 0.5,
-) -> None:
-    """Continuously poll and emit memory-only metrics at a faster cadence.
-
-    Parameters
-    - callback: coroutine called with a fresh MemoryPerformanceProfile each tick
-    - poll_interval_s: interval between polls
-    """
-    while True:
-        try:
-            mem = get_memory_profile()
-            await callback(mem)
-        except MacMonError as e:
-            logger.opt(exception=e).error("Memory Monitor encountered error")
-        finally:
-            await anyio.sleep(poll_interval_s)
-
-
-async def start_polling_node_metrics(
-    callback: Callable[[NodePerformanceProfile], Coroutine[Any, Any, None]],
-):
-    poll_interval_s = 1.0
-    while True:
-        try:
-            metrics = await get_metrics_async()
-            if metrics is None:
-                return
-
-            network_interfaces = get_network_interfaces()
-            # these awaits could be joined but realistically they should be cached
-            model_id, chip_id = await get_model_and_chip()
-            friendly_name = await get_friendly_name()
-
-            # do the memory profile last to get a fresh reading to not conflict with the other memory profiling loop
-            memory_profile = get_memory_profile()
-
-            await callback(
-                NodePerformanceProfile(
-                    model_id=model_id,
-                    chip_id=chip_id,
-                    friendly_name=friendly_name,
-                    network_interfaces=network_interfaces,
-                    memory=memory_profile,
-                    system=SystemPerformanceProfile(
-                        gpu_usage=metrics.gpu_usage[1],
-                        temp=metrics.temp.gpu_temp_avg,
-                        sys_power=metrics.sys_power,
-                        pcpu_usage=metrics.pcpu_usage[1],
-                        ecpu_usage=metrics.ecpu_usage[1],
-                        ane_power=metrics.ane_power,
-                    ),
-                )
-            )
-
-        except asyncio.TimeoutError:
-            logger.warning(
-                "[resource_monitor] Operation timed out after 30s, skipping this cycle."
-            )
-        except MacMonError as e:
-            logger.opt(exception=e).error("Resource Monitor encountered error")
-            return
-        finally:
-            await anyio.sleep(poll_interval_s)

src/exo/worker/utils/tests/test_macmon.py

@@ -1,77 +0,0 @@
-"""Tests for macmon error handling.
-
-These tests verify that MacMon errors are handled gracefully without
-crashing the application or spamming logs.
-"""
-
-import platform
-from subprocess import CalledProcessError
-from unittest.mock import AsyncMock, patch
-
-import pytest
-
-from exo.worker.utils.macmon import MacMonError, get_metrics_async
-
-
-@pytest.mark.skipif(
-    platform.system().lower() != "darwin" or "arm" not in platform.machine().lower(),
-    reason="MacMon only supports macOS with Apple Silicon",
-)
-class TestMacMonErrorHandling:
-    """Test MacMon error handling."""
-
-    async def test_called_process_error_wrapped_as_macmon_error(self) -> None:
-        """CalledProcessError should be wrapped as MacMonError."""
-        mock_error = CalledProcessError(
-            returncode=1,
-            cmd=["macmon", "pipe", "-s", "1"],
-            stderr=b"some error message",
-        )
-
-        with (
-            patch(
-                "exo.worker.utils.macmon.shutil.which", return_value="/usr/bin/macmon"
-            ),
-            patch(
-                "exo.worker.utils.macmon.run_process", new_callable=AsyncMock
-            ) as mock_run,
-        ):
-            mock_run.side_effect = mock_error
-
-            with pytest.raises(MacMonError) as exc_info:
-                await get_metrics_async()
-
-            assert "MacMon failed with return code 1" in str(exc_info.value)
-            assert "some error message" in str(exc_info.value)
-
-    async def test_called_process_error_with_no_stderr(self) -> None:
-        """CalledProcessError with no stderr should be handled gracefully."""
-        mock_error = CalledProcessError(
-            returncode=1,
-            cmd=["macmon", "pipe", "-s", "1"],
-            stderr=None,
-        )
-
-        with (
-            patch(
-                "exo.worker.utils.macmon.shutil.which", return_value="/usr/bin/macmon"
-            ),
-            patch(
-                "exo.worker.utils.macmon.run_process", new_callable=AsyncMock
-            ) as mock_run,
-        ):
-            mock_run.side_effect = mock_error
-
-            with pytest.raises(MacMonError) as exc_info:
-                await get_metrics_async()
-
-            assert "MacMon failed with return code 1" in str(exc_info.value)
-            assert "no stderr" in str(exc_info.value)
-
-    async def test_macmon_not_found_raises_macmon_error(self) -> None:
-        """When macmon is not found in PATH, MacMonError should be raised."""
-        with patch("exo.worker.utils.macmon.shutil.which", return_value=None):
-            with pytest.raises(MacMonError) as exc_info:
-                await get_metrics_async()
-
-            assert "MacMon not found in PATH" in str(exc_info.value)

tests/headless_runner.py

@@ -34,7 +34,8 @@ from exo.shared.types.worker.instances import (
 )
 from exo.shared.types.worker.runners import RunnerId, ShardAssignments
 from exo.shared.types.worker.shards import PipelineShardMetadata, TensorShardMetadata
-from exo.utils.channels import MpReceiver, MpSender, mp_channel
+from exo.utils.channels import MpReceiver, MpSender, channel, mp_channel
+from exo.utils.info_gatherer.info_gatherer import GatheredInfo, InfoGatherer
 from exo.worker.download.impl_shard_downloader import (
     build_full_shard,
     exo_shard_downloader,
@@ -49,14 +50,12 @@ class Tests(BaseModel):
     kind: typing.Literal["init", "warmup", "inference"]
 
 
-hn = socket.gethostname()
 mp.set_start_method("spawn", force=True)
 logger_setup(None)
 
 
 async def main():
     logger.info("starting cool server majig")
-    logger.info(hn)
     await assert_downloads()
     cfg = Config()
     cfg.bind = "0.0.0.0:52415"
@@ -67,6 +66,7 @@ async def main():
     app = FastAPI()
     app.post("/ring")(ring_backend)
     app.post("/jaccl")(jaccl_backend)
+    app.post("/tb_detection")(tb_detection)
     shutdown = anyio.Event()
     await serve(
         app,  # type: ignore
@@ -78,23 +78,53 @@ async def main():
     shutdown.set()
 
 
+async def tb_detection():
+    send, recv = channel[GatheredInfo]()
+    ig = InfoGatherer(send)
+    with anyio.move_on_after(1):
+        await ig._monitor_system_profiler()  # pyright: ignore[reportPrivateUsage]
+    with recv:
+        return recv.collect()
+
+
 async def assert_downloads():
     sd = exo_shard_downloader()
     # await sd.ensure_shard(await build_full_shard(MODEL_CARDS["qwen3-0.6b"].model_id))
-    await sd.ensure_shard(await build_full_shard(MODEL_CARDS["llama-3.2-1b"].model_id))
+    await sd.ensure_shard(
+        await build_full_shard(MODEL_CARDS["llama-3.1-8b-bf16"].model_id)
+    )
+    await sd.ensure_shard(await build_full_shard(MODEL_CARDS["qwen3-30b"].model_id))
+    await sd.ensure_shard(
+        await build_full_shard(MODEL_CARDS["gpt-oss-120b-MXFP4-Q8"].model_id)
+    )
+    await sd.ensure_shard(
+        await build_full_shard(MODEL_CARDS["gpt-oss-20b-4bit"].model_id)
+    )
+    await sd.ensure_shard(
+        await build_full_shard(MODEL_CARDS["glm-4.7-8bit-gs32"].model_id)
+    )
+    await sd.ensure_shard(
+        await build_full_shard(MODEL_CARDS["minimax-m2.1-8bit"].model_id)
+    )
 
 
 async def ring_backend(test: Tests):
     iid = InstanceId(str(hash(str(test.devs))))
-    return await execute_test(test, ring_instance(test, iid))
+    weird_hn = socket.gethostname()
+    for dev in test.devs:
+        if weird_hn.startswith(dev[0]) or dev[0].startswith(weird_hn):
+            hn = dev[0]
+            break
+    else:
+        raise ValueError(f"{weird_hn} not in {test.devs}")
+    return await execute_test(test, ring_instance(test, iid, hn), hn)
 
 
-def ring_instance(test: Tests, iid: InstanceId) -> Instance:
-    global hn
+def ring_instance(test: Tests, iid: InstanceId, hn: str) -> Instance:
     hbn = [Host(ip="i dont care", port=52416) for _ in test.devs]
     world_size = len(test.devs)
     for i in range(world_size):
-        if hn.startswith(test.devs[i][0]):
+        if test.devs[i][0] == hn:
             hn = test.devs[i][0]
             if i - 1 >= 0:
                 hbn[i - 1] = Host(ip=test.devs[i - 1][1], port=52416)
@@ -102,6 +132,8 @@ def ring_instance(test: Tests, iid: InstanceId) -> Instance:
                 hbn[i + 1] = Host(ip=test.devs[i + 1][1], port=52416)
             hbn[i] = Host(ip="0.0.0.0", port=52416)
             break
+    else:
+        raise ValueError(f"{hn} not in {test.devs}")
 
     meta = MODEL_CARDS[test.model_id].metadata
     instance = MlxRingInstance(
@@ -131,10 +163,10 @@ def ring_instance(test: Tests, iid: InstanceId) -> Instance:
     return instance
 
 
-async def execute_test(test: Tests, instance: Instance):
+async def execute_test(test: Tests, instance: Instance, hn: str):
     world_size = len(test.devs)
     iid = InstanceId(str(hash(str(test.devs))))
-    _handle, recv, send = new_runner(instance)
+    _handle, recv, send = new_runner(instance, hn)
     if world_size > 1:
         send.send(ConnectToGroup(instance_id=iid))
     send.send(LoadModel(instance_id=iid))
@@ -181,21 +213,23 @@ async def execute_test(test: Tests, instance: Instance):
 
 async def jaccl_backend(test: Tests):
     iid = InstanceId(str(hash(str(test.devs))))
-    return await execute_test(test, jaccl_instance(test, iid))
+    weird_hn = socket.gethostname()
+    for dev in test.devs:
+        if weird_hn.startswith(dev[0]) or dev[0].startswith(weird_hn):
+            hn = dev[0]
+            break
+    else:
+        raise ValueError(f"{weird_hn} not in {test.devs}")
+    return await execute_test(test, jaccl_instance(test, iid), hn)
 
 
 def jaccl_instance(test: Tests, iid: InstanceId):
-    global hn
     meta = MODEL_CARDS[test.model_id].metadata
     world_size = len(test.devs)
-    for name, _ in test.devs:
-        if hn.startswith(name):
-            hn = name
-            break
 
     return MlxJacclInstance(
         instance_id=iid,
-        ibv_devices=[[None, "rdma_en3"], ["rdma_en3", None]],
+        jaccl_devices=[[None, "rdma_en3"], ["rdma_en3", None]],
         # rank 0 is always coordinator
         jaccl_coordinators={
             NodeId(host[0]): test.devs[0][1] + ":52416" for host in test.devs
@@ -220,6 +254,7 @@ def jaccl_instance(test: Tests, iid: InstanceId):
 
 def new_runner(
     instance: Instance,
+    hn: str,
 ) -> tuple[mp.Process, MpReceiver[Event], MpSender[Task]]:
     bound_instance = BoundInstance(
         instance=instance, bound_runner_id=RunnerId(hn), bound_node_id=NodeId(hn)