woahg

.mlx_typings/mlx_lm/models/glm_moe_dsa.pyi

@@ -1,46 +0,0 @@
-"""Type stubs for mlx_lm.models.glm_moe_dsa"""
-
-from dataclasses import dataclass
-from typing import Any, Dict, Optional
-
-from .base import BaseModelArgs
-from .deepseek_v32 import Model as DSV32Model
-
-@dataclass
-class ModelArgs(BaseModelArgs):
-    model_type: str
-    vocab_size: int
-    hidden_size: int
-    index_head_dim: int
-    index_n_heads: int
-    index_topk: 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: 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_parameters: Dict[str, Any]
-    attention_bias: bool
-    rope_scaling: Dict[str, Any] | None
-    rope_theta: float | None
-
-class Model(DSV32Model):
-    def __init__(self, config: ModelArgs) -> None: ...

Cargo.lock

@@ -877,8 +877,8 @@ name = "exo_pyo3_bindings"
 version = "0.0.1"
 dependencies = [
  "env_logger",
- "extend",
  "futures-lite",
+ "libp2p",
  "log",
  "networking",
  "pyo3",
@@ -889,15 +889,10 @@ dependencies = [
 ]
 
 [[package]]
-name = "extend"
-version = "1.2.0"
+name = "fastrand"
+version = "2.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "311a6d2f1f9d60bff73d2c78a0af97ed27f79672f15c238192a5bbb64db56d00"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn 2.0.111",
-]
+checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be"
 
 [[package]]
 name = "ff"
@@ -1007,7 +1002,10 @@ version = "2.6.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f78e10609fe0e0b3f4157ffab1876319b5b0db102a2c60dc4626306dc46b44ad"
 dependencies = [
+ "fastrand",
  "futures-core",
+ "futures-io",
+ "parking",
  "pin-project-lite",
 ]
 
@@ -1723,12 +1721,6 @@ dependencies = [
  "cpufeatures",
 ]
 
-[[package]]
-name = "keccak-const"
-version = "0.2.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "57d8d8ce877200136358e0bbff3a77965875db3af755a11e1fa6b1b3e2df13ea"
-
 [[package]]
 name = "lalrpop-util"
 version = "0.20.2"
@@ -2735,9 +2727,6 @@ dependencies = [
 name = "networking"
 version = "0.0.1"
 dependencies = [
- "extend",
- "futures-lite",
- "keccak-const",
  "libp2p",
  "log",
  "tokio",

Cargo.toml

@@ -24,32 +24,17 @@ opt-level = 3
 ## Crate members as common dependencies
 networking = { path = "rust/networking" }
 
-# Macro dependecies
-extend = "1.2"
-delegate = "0.13"
-pin-project = "1"
-
-# Utility dependencies
-keccak-const = "0.2"
-
 # Async dependencies
 tokio = "1.46"
-futures = "0.3"
-futures-timer = "3.0"
-
-# Data structures
-either = "1.15"
 
 # Tracing/logging
 log = "0.4"
 
 # networking
 libp2p = "0.56"
-libp2p-tcp = "0.44"
 
 [workspace.lints.rust]
-static_mut_refs = "warn"      # Or use "warn" instead of deny
-incomplete_features = "allow"
+static_mut_refs = "warn"
 
 # Clippy's lint category level configurations;
 # every member crate needs to inherit these by adding
@@ -70,64 +55,3 @@ perf = { level = "warn", priority = -1 }
 pedantic = { level = "warn", priority = -1 }
 nursery = { level = "warn", priority = -1 }
 cargo = { level = "warn", priority = -1 }
-
-# Individual Clippy lints from the `restriction` category
-arithmetic_side_effects = "warn"
-as_conversions = "warn"
-assertions_on_result_states = "warn"
-clone_on_ref_ptr = "warn"
-decimal_literal_representation = "warn"
-default_union_representation = "warn"
-deref_by_slicing = "warn"
-disallowed_script_idents = "deny"
-else_if_without_else = "warn"
-empty_enum_variants_with_brackets = "warn"
-empty_structs_with_brackets = "warn"
-error_impl_error = "warn"
-exit = "deny"
-expect_used = "warn"
-float_cmp_const = "warn"
-get_unwrap = "warn"
-if_then_some_else_none = "warn"
-impl_trait_in_params = "warn"
-indexing_slicing = "warn"
-infinite_loop = "warn"
-let_underscore_must_use = "warn"
-let_underscore_untyped = "warn"
-lossy_float_literal = "warn"
-mem_forget = "warn"
-missing_inline_in_public_items = "warn"
-multiple_inherent_impl = "warn"
-multiple_unsafe_ops_per_block = "warn"
-mutex_atomic = "warn"
-non_zero_suggestions = "warn"
-panic = "warn"
-partial_pub_fields = "warn"
-pattern_type_mismatch = "warn"
-pub_without_shorthand = "warn"
-rc_buffer = "warn"
-rc_mutex = "warn"
-redundant_type_annotations = "warn"
-renamed_function_params = "warn"
-rest_pat_in_fully_bound_structs = "warn"
-same_name_method = "warn"
-self_named_module_files = "deny"
-semicolon_inside_block = "warn"
-shadow_same = "warn"
-shadow_unrelated = "warn"
-str_to_string = "warn"
-string_add = "warn"
-string_lit_chars_any = "warn"
-string_to_string = "warn"
-tests_outside_test_module = "warn"
-todo = "warn"
-try_err = "warn"
-undocumented_unsafe_blocks = "warn"
-unnecessary_safety_comment = "warn"
-unnecessary_safety_doc = "warn"
-unneeded_field_pattern = "warn"
-unseparated_literal_suffix = "warn"
-unused_result_ok = "warn"
-unused_trait_names = "warn"
-unwrap_used = "warn"
-verbose_file_reads = "warn"

MISSED_THINGS.md

@@ -1,5 +1,5 @@
 # Missed things
-[X] Log EXO_LIBP2P_NAMESPACE on start in exo/main.py
+[X] Log 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

README.md

@@ -72,23 +72,16 @@ There are two ways to run exo:
 
 ### Run from Source (macOS)
 
-If you have [Nix](https://nixos.org/) installed, you can skip most of the steps below and run exo directly (after accepting the Cachix cache):
-
-```bash
-nix run .#exo
-```
-
 **Prerequisites:**
-- [Xcode](https://developer.apple.com/xcode/) (provides the Metal ToolChain required for MLX compilation)
 - [brew](https://github.com/Homebrew/brew) (for simple package management on macOS)
-
+  
   ```bash
   /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
   ```
 - [uv](https://github.com/astral-sh/uv) (for Python dependency management)
 - [macmon](https://github.com/vladkens/macmon) (for hardware monitoring on Apple Silicon)
 - [node](https://github.com/nodejs/node) (for building the dashboard)
-
+  
   ```bash
   brew install uv macmon node
   ```
@@ -206,14 +199,14 @@ The app will ask for permission to modify system settings and install a new Netw
 
 **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:
+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 `--namespace` cli arg:
 
 - **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)
+- **Configuration**: Access this setting in the app's Advanced settings (or set the `--namespace` argument when running from source)
 
 The namespace is logged on startup for debugging purposes.
 
@@ -425,4 +418,4 @@ 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.

app/EXO/EXO/ExoProcessController.swift

@@ -82,6 +82,7 @@ final class ExoProcessController: ObservableObject {
 
             let child = Process()
             child.executableURL = executableURL
+            child.arguments = ["--namespace", computeNamespace()]
             let exoHomeURL = Self.exoDirectoryURL
             try? FileManager.default.createDirectory(
                 at: exoHomeURL, withIntermediateDirectories: true
@@ -126,37 +127,11 @@ final class ExoProcessController: ObservableObject {
             return
         }
         process.terminationHandler = nil
-        status = .stopped
-
-        guard process.isRunning else {
-            self.process = nil
-            return
+        if process.isRunning {
+            process.terminate()
         }
-
-        let proc = process
         self.process = nil
-
-        Task.detached {
-            proc.interrupt()
-
-            for _ in 0..<50 {
-                if !proc.isRunning { return }
-                try? await Task.sleep(nanoseconds: 100_000_000)
-            }
-
-            if proc.isRunning {
-                proc.terminate()
-            }
-
-            for _ in 0..<30 {
-                if !proc.isRunning { return }
-                try? await Task.sleep(nanoseconds: 100_000_000)
-            }
-
-            if proc.isRunning {
-                kill(proc.processIdentifier, SIGKILL)
-            }
-        }
+        status = .stopped
     }
 
     func restart() {
@@ -242,7 +217,6 @@ final class ExoProcessController: ObservableObject {
     private func makeEnvironment(for runtimeURL: URL) -> [String: String] {
         var environment = ProcessInfo.processInfo.environment
         environment["EXO_RUNTIME_DIR"] = runtimeURL.path
-        environment["EXO_LIBP2P_NAMESPACE"] = computeNamespace()
         if !hfToken.isEmpty {
             environment["HF_TOKEN"] = hfToken
         }

dashboard/src/lib/components/ChatForm.svelte

@@ -103,7 +103,7 @@
   const modelSupportsThinking = $derived(() => {
     if (!currentModel) return false;
     const caps = modelCapabilities[currentModel] || [];
-    return caps.includes("thinking_toggle") && caps.includes("text");
+    return caps.includes("thinking") && caps.includes("text");
   });
 
   const isEditOnlyWithoutImage = $derived(

dashboard/src/lib/components/ImageParamsPanel.svelte

@@ -59,14 +59,13 @@
   }
 
   const sizeOptions: ImageGenerationParams["size"][] = [
-    "auto",
     "512x512",
     "768x768",
     "1024x1024",
     "1024x768",
     "768x1024",
-    "1024x1536",
-    "1536x1024",
+    "1024x1365",
+    "1365x1024",
   ];
 
   const qualityOptions: ImageGenerationParams["quality"][] = [
@@ -177,90 +176,92 @@
 <div class="border-b border-exo-medium-gray/30 px-3 py-2">
   <!-- Basic params row -->
   <div class="flex items-center gap-3 flex-wrap">
-    <!-- Size -->
-    <div class="flex items-center gap-1.5">
-      <span class="text-xs text-exo-light-gray uppercase tracking-wider"
-        >SIZE:</span
-      >
-      <div class="relative">
-        <button
-          bind:this={sizeButtonRef}
-          type="button"
-          onclick={() => (isSizeDropdownOpen = !isSizeDropdownOpen)}
-          class="bg-exo-medium-gray/50 border border-exo-yellow/30 rounded pl-2 pr-6 py-1 text-xs font-mono text-exo-yellow cursor-pointer transition-all duration-200 hover:border-exo-yellow/50 focus:outline-none focus:border-exo-yellow/70 {isSizeDropdownOpen
-            ? 'border-exo-yellow/70'
-            : ''}"
+    <!-- Size (hidden in edit mode - output size comes from input image) -->
+    {#if !isEditMode}
+      <div class="flex items-center gap-1.5">
+        <span class="text-xs text-exo-light-gray uppercase tracking-wider"
+          >SIZE:</span
         >
-          {params.size.toUpperCase()}
-        </button>
-        <div
-          class="absolute right-1.5 top-1/2 -translate-y-1/2 pointer-events-none transition-transform duration-200 {isSizeDropdownOpen
-            ? 'rotate-180'
-            : ''}"
-        >
-          <svg
-            class="w-3 h-3 text-exo-yellow/60"
-            fill="none"
-            viewBox="0 0 24 24"
-            stroke="currentColor"
+        <div class="relative">
+          <button
+            bind:this={sizeButtonRef}
+            type="button"
+            onclick={() => (isSizeDropdownOpen = !isSizeDropdownOpen)}
+            class="bg-exo-medium-gray/50 border border-exo-yellow/30 rounded pl-2 pr-6 py-1 text-xs font-mono text-exo-yellow cursor-pointer transition-all duration-200 hover:border-exo-yellow/50 focus:outline-none focus:border-exo-yellow/70 {isSizeDropdownOpen
+              ? 'border-exo-yellow/70'
+              : ''}"
           >
-            <path
-              stroke-linecap="round"
-              stroke-linejoin="round"
-              stroke-width="2"
-              d="M19 9l-7 7-7-7"
-            />
-          </svg>
-        </div>
-      </div>
-
-      {#if isSizeDropdownOpen}
-        <!-- Backdrop to close dropdown -->
-        <button
-          type="button"
-          class="fixed inset-0 z-[9998] cursor-default"
-          onclick={() => (isSizeDropdownOpen = false)}
-          aria-label="Close dropdown"
-        ></button>
-
-        <!-- Dropdown Panel - fixed positioning to escape overflow:hidden -->
-        <div
-          class="fixed bg-exo-dark-gray border border-exo-yellow/30 rounded shadow-lg shadow-black/50 z-[9999] max-h-48 overflow-y-auto overflow-x-hidden min-w-max"
-          style="bottom: calc(100vh - {sizeDropdownPosition()
-            .top}px + 4px); left: {sizeDropdownPosition().left}px;"
-        >
-          <div class="py-1">
-            {#each sizeOptions as size}
-              <button
-                type="button"
-                onclick={() => selectSize(size)}
-                class="w-full px-3 py-1.5 text-left text-xs font-mono tracking-wide transition-colors duration-100 flex items-center gap-2 {params.size ===
-                size
-                  ? 'bg-transparent text-exo-yellow'
-                  : 'text-exo-light-gray hover:text-exo-yellow'}"
-              >
-                {#if params.size === size}
-                  <svg
-                    class="w-3 h-3 flex-shrink-0"
-                    fill="currentColor"
-                    viewBox="0 0 20 20"
-                  >
-                    <path
-                      fill-rule="evenodd"
-                      d="M16.707 5.293a1 1 0 010 1.414l-8 8a1 1 0 01-1.414 0l-4-4a1 1 0 011.414-1.414L8 12.586l7.293-7.293a1 1 0 011.414 0z"
-                      clip-rule="evenodd"
-                    />
-                  </svg>
-                {:else}
-                  <span class="w-3"></span>
-                {/if}
-                <span>{size.toUpperCase()}</span>
-              </button>
-            {/each}
+            {params.size}
+          </button>
+          <div
+            class="absolute right-1.5 top-1/2 -translate-y-1/2 pointer-events-none transition-transform duration-200 {isSizeDropdownOpen
+              ? 'rotate-180'
+              : ''}"
+          >
+            <svg
+              class="w-3 h-3 text-exo-yellow/60"
+              fill="none"
+              viewBox="0 0 24 24"
+              stroke="currentColor"
+            >
+              <path
+                stroke-linecap="round"
+                stroke-linejoin="round"
+                stroke-width="2"
+                d="M19 9l-7 7-7-7"
+              />
+            </svg>
           </div>
         </div>
-      {/if}
-    </div>
+
+        {#if isSizeDropdownOpen}
+          <!-- Backdrop to close dropdown -->
+          <button
+            type="button"
+            class="fixed inset-0 z-[9998] cursor-default"
+            onclick={() => (isSizeDropdownOpen = false)}
+            aria-label="Close dropdown"
+          ></button>
+
+          <!-- Dropdown Panel - fixed positioning to escape overflow:hidden -->
+          <div
+            class="fixed bg-exo-dark-gray border border-exo-yellow/30 rounded shadow-lg shadow-black/50 z-[9999] max-h-48 overflow-y-auto min-w-max"
+            style="bottom: calc(100vh - {sizeDropdownPosition()
+              .top}px + 4px); left: {sizeDropdownPosition().left}px;"
+          >
+            <div class="py-1">
+              {#each sizeOptions as size}
+                <button
+                  type="button"
+                  onclick={() => selectSize(size)}
+                  class="w-full px-3 py-1.5 text-left text-xs font-mono tracking-wide transition-colors duration-100 flex items-center gap-2 {params.size ===
+                  size
+                    ? 'bg-transparent text-exo-yellow'
+                    : 'text-exo-light-gray hover:text-exo-yellow'}"
+                >
+                  {#if params.size === size}
+                    <svg
+                      class="w-3 h-3 flex-shrink-0"
+                      fill="currentColor"
+                      viewBox="0 0 20 20"
+                    >
+                      <path
+                        fill-rule="evenodd"
+                        d="M16.707 5.293a1 1 0 010 1.414l-8 8a1 1 0 01-1.414 0l-4-4a1 1 0 011.414-1.414L8 12.586l7.293-7.293a1 1 0 011.414 0z"
+                        clip-rule="evenodd"
+                      />
+                    </svg>
+                  {:else}
+                    <span class="w-3"></span>
+                  {/if}
+                  <span>{size}</span>
+                </button>
+              {/each}
+            </div>
+          </div>
+        {/if}
+      </div>
+    {/if}
 
     <!-- Quality -->
     <div class="flex items-center gap-1.5">
@@ -310,7 +311,7 @@
 
         <!-- Dropdown Panel - fixed positioning to escape overflow:hidden -->
         <div
-          class="fixed bg-exo-dark-gray border border-exo-yellow/30 rounded shadow-lg shadow-black/50 z-[9999] max-h-48 overflow-y-auto overflow-x-hidden min-w-max"
+          class="fixed bg-exo-dark-gray border border-exo-yellow/30 rounded shadow-lg shadow-black/50 z-[9999] max-h-48 overflow-y-auto min-w-max"
           style="bottom: calc(100vh - {qualityDropdownPosition()
             .top}px + 4px); left: {qualityDropdownPosition().left}px;"
         >

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

@@ -306,14 +306,13 @@ const IMAGE_PARAMS_STORAGE_KEY = "exo-image-generation-params";
 export interface ImageGenerationParams {
   // Basic params
   size:
-    | "auto"
     | "512x512"
     | "768x768"
     | "1024x1024"
     | "1024x768"
     | "768x1024"
-    | "1024x1536"
-    | "1536x1024";
+    | "1024x1365"
+    | "1365x1024";
   quality: "low" | "medium" | "high";
   outputFormat: "png" | "jpeg";
   numImages: number;
@@ -337,7 +336,7 @@ export interface EditingImage {
 }
 
 const DEFAULT_IMAGE_PARAMS: ImageGenerationParams = {
-  size: "auto",
+  size: "1024x1024",
   quality: "medium",
   outputFormat: "png",
   numImages: 1,

flake.nix

@@ -115,7 +115,7 @@
           packages = lib.optionalAttrs pkgs.stdenv.hostPlatform.isDarwin (
             let
               uvLock = builtins.fromTOML (builtins.readFile ./uv.lock);
-              mlxPackage = builtins.head (builtins.filter (p: p.name == "mlx" && p.source ? git) uvLock.package);
+              mlxPackage = builtins.head (builtins.filter (p: p.name == "mlx") uvLock.package);
               uvLockMlxVersion = mlxPackage.version;
             in
             {

nix/mlx.nix

@@ -41,16 +41,16 @@ let
 
   mlx = stdenv.mkDerivation rec {
     pname = "mlx";
-    version = let v = "0.30.7.dev20260218+14841977"; in
+    version = let v = "0.30.6"; in
       assert v == uvLockMlxVersion || throw "MLX version mismatch: nix/mlx.nix has ${v} but uv.lock has ${uvLockMlxVersion}. Update both the version and hash in nix/mlx.nix.";
       v;
     pyproject = true;
 
     src = fetchFromGitHub {
-      owner = "rltakashige";
-      repo = "mlx-jaccl-fix-small-recv";
-      rev = "1484197707f35186ad3bd614357c7c47fdf86ebc";
-      hash = "sha256-FupCMoK/SF/ldfKuvMSAKECcOP8c+ANgkQlPZttDsLk=";
+      owner = "ml-explore";
+      repo = "mlx";
+      tag = "v${version}";
+      hash = "sha256-avD5EGhwgmPdXLAyQSqTO6AXk/W3ziH+f6AetjK3Sdo=";
     };
 
     patches = [

pyproject.toml

@@ -17,9 +17,9 @@ dependencies = [
     "loguru>=0.7.3",
     "exo_pyo3_bindings", # rust bindings
     "anyio==4.11.0",
-    "mlx; sys_platform == 'darwin'",
+    "mlx==0.30.6; sys_platform == 'darwin'",
     "mlx[cpu]==0.30.6; sys_platform == 'linux'",
-    "mlx-lm==0.30.7",
+    "mlx-lm==0.30.6",
     "tiktoken>=0.12.0", # required for kimi k2 tokenizer
     "hypercorn>=0.18.0",
     "openai-harmony>=0.0.8",
@@ -64,7 +64,6 @@ members = [
 
 [tool.uv.sources]
 exo_pyo3_bindings = { workspace = true }
-mlx = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git", branch = "address-rdma-gpu-locks", marker = "sys_platform == 'darwin'" }
 #mlx-lm = { git = "https://github.com/davidmcc73/mlx-lm", branch = "stable" }
 # Uncomment to use local mlx/mlx-lm development versions:
 # mlx = { path = "/Users/Shared/mlx", editable=true }
@@ -133,7 +132,7 @@ markers = [
 env = [
   "EXO_TESTS=1"
 ]
-addopts = "-m 'not slow' --ignore=tests/start_distributed_test.py"
+addopts = "-m 'not slow'"
 filterwarnings = [
     "ignore:builtin type Swig:DeprecationWarning",
 ]

python/parts.nix

@@ -58,21 +58,6 @@
         lib.optionalAttrs pkgs.stdenv.hostPlatform.isLinux (
           (lib.mapAttrs (_: ignoreMissing) nvidiaPackages) // {
             mlx = ignoreMissing prev.mlx;
-            mlx-cuda-13 = prev.mlx-cuda-13.overrideAttrs (old: {
-              buildInputs = (old.buildInputs or [ ]) ++ [
-                final.nvidia-cublas
-                final.nvidia-cuda-nvrtc
-                final.nvidia-cudnn-cu13
-                final.nvidia-nccl-cu13
-              ];
-              preFixup = ''
-                addAutoPatchelfSearchPath ${final.nvidia-cublas}
-                addAutoPatchelfSearchPath ${final.nvidia-cuda-nvrtc}
-                addAutoPatchelfSearchPath ${final.nvidia-cudnn-cu13}
-                addAutoPatchelfSearchPath ${final.nvidia-nccl-cu13}
-              '';
-              autoPatchelfIgnoreMissingDeps = [ "libcuda.so.1" ];
-            });
             torch = ignoreMissing prev.torch;
             triton = ignoreMissing prev.triton;
           }
@@ -89,25 +74,14 @@
           linuxOverlay
         ]
       );
-      # mlx-cpu and mlx-cuda-13 both ship mlx/ site-packages files; keep first.
-      # mlx-cpu/mlx-cuda-13 and nvidia-cudnn-cu12/cu13 ship overlapping files.
-      venvCollisionPaths = lib.optionals pkgs.stdenv.hostPlatform.isLinux [
-        "lib/python3.13/site-packages/mlx*"
-        "lib/python3.13/site-packages/nvidia*"
-      ];
-
-      exoVenv = (pythonSet.mkVirtualEnv "exo-env" workspace.deps.default).overrideAttrs {
-        venvIgnoreCollisions = venvCollisionPaths;
-      };
+      exoVenv = pythonSet.mkVirtualEnv "exo-env" workspace.deps.default;
 
       # Virtual environment with dev dependencies for testing
-      testVenv = (pythonSet.mkVirtualEnv "exo-test-env" (
+      testVenv = pythonSet.mkVirtualEnv "exo-test-env" (
         workspace.deps.default // {
           exo = [ "dev" ]; # Include pytest, pytest-asyncio, pytest-env
         }
-      )).overrideAttrs {
-        venvIgnoreCollisions = venvCollisionPaths;
-      };
+      );
 
       mkPythonScript = name: path: pkgs.writeShellApplication {
         inherit name;

resources/inference_model_cards/mlx-community--DeepSeek-V3.1-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "deepseek"
 quantization = "4bit"
 base_model = "DeepSeek V3.1"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 405874409472

resources/inference_model_cards/mlx-community--DeepSeek-V3.1-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "deepseek"
 quantization = "8bit"
 base_model = "DeepSeek V3.1"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 765577920512

resources/inference_model_cards/mlx-community--GLM-4.5-Air-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "8bit"
 base_model = "GLM 4.5 Air"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 122406567936

resources/inference_model_cards/mlx-community--GLM-4.5-Air-bf16.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "bf16"
 base_model = "GLM 4.5 Air"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 229780750336

resources/inference_model_cards/mlx-community--GLM-4.7-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "4bit"
 base_model = "GLM 4.7"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 198556925568

resources/inference_model_cards/mlx-community--GLM-4.7-6bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "6bit"
 base_model = "GLM 4.7"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 286737579648

resources/inference_model_cards/mlx-community--GLM-4.7-8bit-gs32.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "8bit"
 base_model = "GLM 4.7"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 396963397248

resources/inference_model_cards/mlx-community--GLM-4.7-Flash-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "4bit"
 base_model = "GLM 4.7 Flash"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 19327352832

resources/inference_model_cards/mlx-community--GLM-4.7-Flash-5bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "5bit"
 base_model = "GLM 4.7 Flash"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 22548578304

resources/inference_model_cards/mlx-community--GLM-4.7-Flash-6bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "6bit"
 base_model = "GLM 4.7 Flash"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 26843545600

resources/inference_model_cards/mlx-community--GLM-4.7-Flash-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "glm"
 quantization = "8bit"
 base_model = "GLM 4.7 Flash"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 34359738368

resources/inference_model_cards/mlx-community--GLM-5-8bit.toml

@@ -1,12 +0,0 @@
-model_id = "mlx-community/GLM-5-8bit-MXFP8"
-n_layers = 78
-hidden_size = 6144
-supports_tensor = true
-tasks = ["TextGeneration"]
-family = "glm"
-quantization = "8bit"
-base_model = "GLM-5"
-capabilities = ["text", "thinking"]
-
-[storage_size]
-in_bytes = 790517400864

resources/inference_model_cards/mlx-community--GLM-5-MXFP4-Q8.toml

@@ -1,12 +0,0 @@
-model_id = "mlx-community/GLM-5-MXFP4-Q8"
-n_layers = 78
-hidden_size = 6144
-supports_tensor = true
-tasks = ["TextGeneration"]
-family = "glm"
-quantization = "MXFP4-Q8"
-base_model = "GLM-5"
-capabilities = ["text", "thinking"]
-
-[storage_size]
-in_bytes = 405478939008

resources/inference_model_cards/mlx-community--GLM-5-bf16.toml

@@ -1,12 +0,0 @@
-model_id = "mlx-community/GLM-5"
-n_layers = 78
-hidden_size = 6144
-supports_tensor = true
-tasks = ["TextGeneration"]
-family = "glm"
-quantization = "bf16"
-base_model = "GLM-5"
-capabilities = ["text", "thinking"]
-
-[storage_size]
-in_bytes = 1487822475264

resources/inference_model_cards/mlx-community--Kimi-K2-Thinking.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "kimi"
 quantization = ""
 base_model = "Kimi K2"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 706522120192

resources/inference_model_cards/mlx-community--Kimi-K2.5.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "kimi"
 quantization = ""
 base_model = "Kimi K2.5"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 662498705408

resources/inference_model_cards/mlx-community--MiniMax-M2.1-3bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "minimax"
 quantization = "3bit"
 base_model = "MiniMax M2.1"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 100086644736

resources/inference_model_cards/mlx-community--MiniMax-M2.1-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "minimax"
 quantization = "8bit"
 base_model = "MiniMax M2.1"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 242986745856

resources/inference_model_cards/mlx-community--Qwen3-0.6B-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "4bit"
 base_model = "Qwen3 0.6B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 342884352

resources/inference_model_cards/mlx-community--Qwen3-0.6B-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "8bit"
 base_model = "Qwen3 0.6B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 698351616

resources/inference_model_cards/mlx-community--Qwen3-235B-A22B-Instruct-2507-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "4bit"
 base_model = "Qwen3 235B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 141733920768

resources/inference_model_cards/mlx-community--Qwen3-235B-A22B-Instruct-2507-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "8bit"
 base_model = "Qwen3 235B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 268435456000

resources/inference_model_cards/mlx-community--Qwen3-30B-A3B-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "4bit"
 base_model = "Qwen3 30B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 17612931072

resources/inference_model_cards/mlx-community--Qwen3-30B-A3B-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "8bit"
 base_model = "Qwen3 30B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 33279705088

resources/inference_model_cards/mlx-community--Qwen3-Next-80B-A3B-Thinking-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "4bit"
 base_model = "Qwen3 Next 80B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 47080074240

resources/inference_model_cards/mlx-community--Qwen3-Next-80B-A3B-Thinking-8bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "qwen"
 quantization = "8bit"
 base_model = "Qwen3 Next 80B"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 88814387200

resources/inference_model_cards/mlx-community--Step-3.5-Flash-4bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "step"
 quantization = "4bit"
 base_model = "Step 3.5 Flash"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 114572190076

resources/inference_model_cards/mlx-community--Step-3.5-Flash-6bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "step"
 quantization = "6bit"
 base_model = "Step 3.5 Flash"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 159039627774

resources/inference_model_cards/mlx-community--Step-3.5-Flash-8Bit.toml

@@ -6,7 +6,7 @@ tasks = ["TextGeneration"]
 family = "step"
 quantization = "8bit"
 base_model = "Step 3.5 Flash"
-capabilities = ["text", "thinking", "thinking_toggle"]
+capabilities = ["text", "thinking"]
 
 [storage_size]
 in_bytes = 209082699847

rust/exo_pyo3_bindings/Cargo.toml

@@ -5,6 +5,7 @@ edition = { workspace = true }
 publish = false
 
 [lib]
+doctest = false
 path = "src/lib.rs"
 name = "exo_pyo3_bindings"
 
@@ -25,11 +26,11 @@ networking = { workspace = true }
 
 # interop
 pyo3 = { version = "0.27.2", features = [
-    # "abi3-py313", # tells pyo3 (and maturin) to build using the stable ABI with minimum Python version 3.13
-    "nightly", # enables better-supported GIL integration
-    "experimental-async", # async support in #[pyfunction] & #[pymethods]
-    #"experimental-inspect", # inspection of generated binary => easier to automate type-hint generation
-    #"py-clone", # adding Clone-ing of `Py<T>` without GIL (may cause panics - remove if panics happen)
+    "abi3-py313", # tells pyo3 (and maturin) to build using the stable ABI with minimum Python version 3.13
+    # "nightly", # enables better-supported GIL integration
+    "experimental-async" # async support in #[pyfunction] & #[pymethods]
+    # "experimental-inspect", # inspection of generated binary => easier to automate type-hint generation
+    # "py-clone", # adding Clone-ing of `Py<T>` without GIL (may cause panics - remove if panics happen)
     # "multiple-pymethods", # allows multiple #[pymethods] sections per class
 
     # integrations with other libraries
@@ -41,13 +42,18 @@ pyo3-stub-gen = { version = "0.17.2" }
 pyo3-async-runtimes = { version = "0.27.0", features = ["attributes", "tokio-runtime", "testing"] }
 pyo3-log = "0.13.2"
 
-# macro dependencies
-extend = { workspace = true }
-
 # async runtime
 tokio = { workspace = true, features = ["full", "tracing"] }
-futures = { workspace = true }
+futures-lite = "2.6.1"
 
 # Tracing
+#tracing = "0.1"
+#tracing-subscriber = "0.3"
+#console-subscriber = "0.1.5"
+#tracing-log = "0.2.0"
 log = { workspace = true }
 env_logger = "0.11"
+
+
+# Networking
+libp2p = { workspace = true, features = ["full"] }

rust/exo_pyo3_bindings/README.md

@@ -0,0 +1 @@
+TODO: do something here....

rust/exo_pyo3_bindings/exo_pyo3_bindings.pyi

@@ -2,220 +2,39 @@
 # ruff: noqa: E501, F401
 
 import builtins
-import enum
 import typing
 
-@typing.final
-class AllQueuesFullError(builtins.Exception):
-    def __new__(cls, *args: typing.Any) -> AllQueuesFullError: ...
-    def __repr__(self) -> builtins.str: ...
-    def __str__(self) -> builtins.str: ...
-
-@typing.final
-class ConnectionUpdate:
-    @property
-    def update_type(self) -> ConnectionUpdateType:
-        r"""
-        Whether this is a connection or disconnection event
-        """
-    @property
-    def peer_id(self) -> PeerId:
-        r"""
-        Identity of the peer that we have connected to or disconnected from.
-        """
-    @property
-    def remote_ipv4(self) -> builtins.str:
-        r"""
-        Remote connection's IPv4 address.
-        """
-    @property
-    def remote_tcp_port(self) -> builtins.int:
-        r"""
-        Remote connection's TCP port.
-        """
-
 @typing.final
 class Keypair:
-    r"""
-    Identity keypair of a node.
-    """
     @staticmethod
-    def generate_ed25519() -> Keypair:
+    def generate() -> Keypair:
         r"""
-        Generate a new Ed25519 keypair.
-        """
-    @staticmethod
-    def generate_ecdsa() -> Keypair:
-        r"""
-        Generate a new ECDSA keypair.
-        """
-    @staticmethod
-    def generate_secp256k1() -> Keypair:
-        r"""
-        Generate a new Secp256k1 keypair.
+        Generate a new ed25519 keypair
         """
     @staticmethod
     def from_protobuf_encoding(bytes: bytes) -> Keypair:
         r"""
         Decode a private key from a protobuf structure and parse it as a `Keypair`.
         """
-    @staticmethod
-    def rsa_from_pkcs8(bytes: bytes) -> Keypair:
-        r"""
-        Decode an keypair from a DER-encoded secret key in PKCS#8 `PrivateKeyInfo`
-        format (i.e. unencrypted) as defined in [RFC5208].
-        
-        [RFC5208]: https://tools.ietf.org/html/rfc5208#section-5
-        """
-    @staticmethod
-    def secp256k1_from_der(bytes: bytes) -> Keypair:
-        r"""
-        Decode a keypair from a DER-encoded Secp256k1 secret key in an `ECPrivateKey`
-        structure as defined in [RFC5915].
-        
-        [RFC5915]: https://tools.ietf.org/html/rfc5915
-        """
-    @staticmethod
-    def ed25519_from_bytes(bytes: bytes) -> Keypair: ...
     def to_protobuf_encoding(self) -> bytes:
         r"""
-        Encode a private key as protobuf structure.
-        """
-    def to_peer_id(self) -> PeerId:
-        r"""
-        Convert the `Keypair` into the corresponding `PeerId`.
+        Encode a private key to a protobuf structure.
         """
+    def to_string(self) -> builtins.str: ...
 
 @typing.final
-class Multiaddr:
-    r"""
-    Representation of a Multiaddr.
-    """
+class PyPeer:
     @staticmethod
-    def empty() -> Multiaddr:
-        r"""
-        Create a new, empty multiaddress.
-        """
-    @staticmethod
-    def with_capacity(n: builtins.int) -> Multiaddr:
-        r"""
-        Create a new, empty multiaddress with the given capacity.
-        """
-    @staticmethod
-    def from_bytes(bytes: bytes) -> Multiaddr:
-        r"""
-        Parse a `Multiaddr` value from its byte slice representation.
-        """
-    @staticmethod
-    def from_string(string: builtins.str) -> Multiaddr:
-        r"""
-        Parse a `Multiaddr` value from its string representation.
-        """
-    def len(self) -> builtins.int:
-        r"""
-        Return the length in bytes of this multiaddress.
-        """
-    def is_empty(self) -> builtins.bool:
-        r"""
-        Returns true if the length of this multiaddress is 0.
-        """
-    def to_bytes(self) -> bytes:
-        r"""
-        Return a copy of this [`Multiaddr`]'s byte representation.
-        """
-    def to_string(self) -> builtins.str:
-        r"""
-        Convert a Multiaddr to a string.
-        """
+    def new(kp: Keypair, namespace: builtins.str) -> PyPeer: ...
+    async def subscribe(self, topic: builtins.str) -> None: ...
+    async def unsubscribe(self, topic: builtins.str) -> None: ...
+    async def send(self, topic: builtins.str, payload: bytes) -> None: ...
+    async def run(self) -> None: ...
+    async def recv(self) -> PySwarmEvent: ...
 
 @typing.final
-class NetworkingHandle:
-    def __new__(cls, identity: Keypair) -> NetworkingHandle: ...
-    async def connection_update_recv(self) -> ConnectionUpdate:
-        r"""
-        Receives the next `ConnectionUpdate` from networking.
-        """
-    async def connection_update_recv_many(self, limit: builtins.int) -> builtins.list[ConnectionUpdate]:
-        r"""
-        Receives at most `limit` `ConnectionUpdate`s from networking and returns them.
-        
-        For `limit = 0`, an empty collection of `ConnectionUpdate`s will be returned immediately.
-        For `limit > 0`, if there are no `ConnectionUpdate`s in the channel's queue this method
-        will sleep until a `ConnectionUpdate`s is sent.
-        """
-    async def gossipsub_subscribe(self, topic: builtins.str) -> builtins.bool:
-        r"""
-        Subscribe to a `GossipSub` topic.
-        
-        Returns `True` if the subscription worked. Returns `False` if we were already subscribed.
-        """
-    async def gossipsub_unsubscribe(self, topic: builtins.str) -> builtins.bool:
-        r"""
-        Unsubscribes from a `GossipSub` topic.
-        
-        Returns `True` if we were subscribed to this topic. Returns `False` if we were not subscribed.
-        """
-    async def gossipsub_publish(self, topic: builtins.str, data: bytes) -> None:
-        r"""
-        Publishes a message with multiple topics to the `GossipSub` network.
-        
-        If no peers are found that subscribe to this topic, throws `NoPeersSubscribedToTopicError` exception.
-        """
-    async def gossipsub_recv(self) -> tuple[builtins.str, bytes]:
-        r"""
-        Receives the next message from the `GossipSub` network.
-        """
-    async def gossipsub_recv_many(self, limit: builtins.int) -> builtins.list[tuple[builtins.str, bytes]]:
-        r"""
-        Receives at most `limit` messages from the `GossipSub` network and returns them.
-        
-        For `limit = 0`, an empty collection of messages will be returned immediately.
-        For `limit > 0`, if there are no messages in the channel's queue this method
-        will sleep until a message is sent.
-        """
-
-@typing.final
-class NoPeersSubscribedToTopicError(builtins.Exception):
-    def __new__(cls, *args: typing.Any) -> NoPeersSubscribedToTopicError: ...
-    def __repr__(self) -> builtins.str: ...
-    def __str__(self) -> builtins.str: ...
-
-@typing.final
-class PeerId:
-    r"""
-    Identifier of a peer of the network.
-    
-    The data is a `CIDv0` compatible multihash of the protobuf encoded public key of the peer
-    as specified in [specs/peer-ids](https://github.com/libp2p/specs/blob/master/peer-ids/peer-ids.md).
-    """
-    @staticmethod
-    def random() -> PeerId:
-        r"""
-        Generates a random peer ID from a cryptographically secure PRNG.
-        
-        This is useful for randomly walking on a DHT, or for testing purposes.
-        """
-    @staticmethod
-    def from_bytes(bytes: bytes) -> PeerId:
-        r"""
-        Parses a `PeerId` from bytes.
-        """
-    def to_bytes(self) -> bytes:
-        r"""
-        Returns a raw bytes representation of this `PeerId`.
-        """
-    def to_base58(self) -> builtins.str:
-        r"""
-        Returns a base-58 encoded string of this `PeerId`.
-        """
-    def __repr__(self) -> builtins.str: ...
-    def __str__(self) -> builtins.str: ...
-
-@typing.final
-class ConnectionUpdateType(enum.Enum):
-    r"""
-    Connection or disconnection event discriminant type.
-    """
-    Connected = ...
-    Disconnected = ...
+class PySwarmEvent:
+    def downcast_discovered(self) -> typing.Optional[builtins.str]: ...
+    def downcast_expired(self) -> typing.Optional[builtins.str]: ...
+    def downcast_message(self) -> typing.Optional[tuple[builtins.str, builtins.str, bytes]]: ...
 

rust/exo_pyo3_bindings/src/lib.rs

@@ -1,5 +1,43 @@
+//! TODO: crate documentation
+pub(crate) mod allow_threading;
+
+pub(crate) mod networking;
+pub(crate) mod take_once {
+    use std::sync::Mutex;
+
+    pub struct TakeOnce<T>(Mutex<Option<T>>);
+    impl<T> TakeOnce<T> {
+        pub fn new(t: T) -> Self {
+            Self(Mutex::new(Some(t)))
+        }
+        pub fn take(&self) -> Option<T> {
+            match self.0.try_lock() {
+                Ok(mut o) => o.take(),
+                Err(_) => None,
+            }
+        }
+    }
+}
+
+use pyo3::prelude::*;
+
 use pyo3_stub_gen::define_stub_info_gatherer;
 
-mod allow_threading;
+/// A Python module implemented in Rust. The name of this function must match
+/// the `lib.name` setting in the `Cargo.toml`, else Python will not be able to
+/// import the module.
+#[pymodule(name = "exo_pyo3_bindings")]
+pub fn networking_module(m: &Bound<'_, PyModule>) -> PyResult<()> {
+    // install logger
+    pyo3_log::init();
+    // setup runtime
+    let mut builder = tokio::runtime::Builder::new_multi_thread();
+    builder.enable_all();
+    pyo3_async_runtimes::tokio::init(builder);
+
+    m.add_class::<networking::PyPeer>()?;
+    m.add_class::<networking::PyKeypair>()?;
+    Ok(())
+}
 
 define_stub_info_gatherer!(stub_info);

rust/exo_pyo3_bindings/src/networking.rs

@@ -0,0 +1,214 @@
+use crate::allow_threading::AllowThreads;
+use crate::take_once::TakeOnce;
+
+use std::pin::pin;
+
+use futures_lite::FutureExt;
+use libp2p::{gossipsub::PublishError, identity::Keypair};
+use networking::{FromSwarm, Peer, ToSwarm};
+use pyo3::{
+    coroutine::CancelHandle,
+    exceptions::{PyConnectionError, PyRuntimeError, PyValueError},
+    prelude::*,
+    types::PyBytes,
+};
+use pyo3_stub_gen::{
+    derive::{gen_methods_from_python, gen_stub_pyclass, gen_stub_pymethods},
+    inventory::submit,
+};
+use tokio::sync::{Mutex, mpsc};
+
+#[gen_stub_pyclass]
+#[pyclass(name = "Keypair", frozen)]
+#[derive(Clone)]
+pub struct PyKeypair(Keypair);
+
+#[gen_stub_pymethods]
+#[pymethods]
+impl PyKeypair {
+    /// Generate a new ed25519 keypair
+    #[staticmethod]
+    fn generate() -> Self {
+        Self(Keypair::generate_ed25519())
+    }
+
+    /// Decode a private key from a protobuf structure and parse it as a `Keypair`.
+    #[staticmethod]
+    fn from_protobuf_encoding(bytes: &Bound<'_, PyBytes>) -> Self {
+        let bytes = Vec::from(bytes.as_bytes());
+        Self(Keypair::from_protobuf_encoding(&bytes).expect("todo"))
+    }
+
+    /// Encode a private key to a protobuf structure.
+    fn to_protobuf_encoding<'py>(&self, py: Python<'py>) -> PyResult<Bound<'py, PyBytes>> {
+        match self.0.to_protobuf_encoding() {
+            Ok(bytes) => Ok(PyBytes::new(py, &bytes)),
+            Err(e) => Err(PyValueError::new_err(e.to_string())),
+        }
+    }
+
+    fn to_string(&self) -> String {
+        self.0.public().to_peer_id().to_base58()
+    }
+}
+
+struct PeerBuilder(
+    String,
+    Keypair,
+    mpsc::Sender<FromSwarm>,
+    mpsc::Receiver<ToSwarm>,
+);
+
+#[gen_stub_pyclass]
+#[pyclass]
+pub struct PyPeer {
+    peer: TakeOnce<PeerBuilder>,
+    to_swarm: mpsc::Sender<ToSwarm>,
+    from_swarm: Mutex<mpsc::Receiver<FromSwarm>>,
+}
+
+#[gen_stub_pymethods]
+#[pymethods]
+impl PyPeer {
+    #[staticmethod]
+    fn new(kp: PyKeypair, namespace: String) -> PyResult<Self> {
+        let (to_client, from_swarm) = mpsc::channel(1024);
+        let (to_swarm, from_client) = mpsc::channel(1024);
+        Ok(Self {
+            peer: TakeOnce::new(PeerBuilder(namespace, kp.0, to_client, from_client)),
+            to_swarm,
+            from_swarm: Mutex::new(from_swarm),
+        })
+    }
+
+    #[gen_stub(skip)]
+    async fn run(&self, #[pyo3(cancel_handle)] mut cancel: CancelHandle) -> PyResult<()> {
+        let builder = self
+            .peer
+            .take()
+            .ok_or_else(|| PyRuntimeError::new_err("tried to run peer twice"))?;
+        let jh = pyo3_async_runtimes::tokio::get_runtime()
+            .spawn(async move {
+                let mut peer =
+                    Peer::new(builder.0, builder.1, builder.2, builder.3).map_err(|_| {
+                        PyConnectionError::new_err("peer failed to listen on default address")
+                    })?;
+                peer.run()
+                    .await
+                    .map_err(|()| PyConnectionError::new_err("peer communication closed"))
+            })
+            .or(async {
+                cancel.cancelled().await;
+                Ok(Ok(()))
+            });
+        match AllowThreads(pin!(jh)).await {
+            Err(e) if e.is_cancelled() => Ok(()),
+            Err(e) if e.is_panic() => Err(PyRuntimeError::new_err(format!("tokio panic {e}"))),
+            Err(_) => unreachable!(),
+            Ok(res) => res,
+        }
+    }
+
+    async fn subscribe(&self, topic: String) -> PyResult<()> {
+        self.to_swarm
+            .send(ToSwarm::Subscribe(topic))
+            .await
+            .map_err(|_| PyRuntimeError::new_err("swarm communication closed"))
+    }
+    async fn unsubscribe(&self, topic: String) -> PyResult<()> {
+        self.to_swarm
+            .send(ToSwarm::Unsubscribe(topic))
+            .await
+            .map_err(|_| PyRuntimeError::new_err("swarm communication closed"))
+    }
+    async fn send(&self, topic: String, payload: Py<PyBytes>) -> PyResult<()> {
+        // this function attaches to the python interpreter synchronously to avoid holding the GIL
+        let bytes = Python::attach(|py| Vec::from(payload.bind(py).as_bytes()));
+        self.to_swarm
+            .send(ToSwarm::Message(topic, bytes))
+            .await
+            .map_err(|_| PyRuntimeError::new_err("swarm communication closed"))
+    }
+
+    #[gen_stub(skip)]
+    async fn recv(
+        &self,
+        #[pyo3(cancel_handle)] mut cancel: CancelHandle,
+    ) -> PyResult<PySwarmEvent> {
+        loop {
+            return match AllowThreads(pin!(
+                self.from_swarm
+                    .try_lock()
+                    .map_err(|_| PyRuntimeError::new_err("tried to recv twice"))?
+                    .recv()
+                    .or(async {
+                        cancel.cancelled().await;
+                        None
+                    })
+            ))
+            .await
+            {
+                Some(FromSwarm::PublishError(p)) => match p {
+                    PublishError::AllQueuesFull(_) => {
+                        Err(PyConnectionError::new_err("swarm overloaded"))
+                    }
+                    PublishError::MessageTooLarge => {
+                        Err(PyValueError::new_err("message too large"))
+                    }
+                    PublishError::NoPeersSubscribedToTopic => {
+                        continue;
+                    }
+                    // TODO(evan): logs here
+                    _ => continue,
+                },
+                None => Err(PyRuntimeError::new_err("swarm communication closed")),
+                Some(fs) => Ok(PySwarmEvent(fs)),
+            };
+        }
+    }
+}
+
+// Manually submit the run()/recv() stub because the cancelhandle is poorly understood
+submit! {
+    gen_methods_from_python! {
+        r#"
+        class PyPeer:
+            async def run(self): ...
+            async def recv(self) -> PySwarmEvent: ...
+        "#
+    }
+}
+
+#[gen_stub_pyclass]
+#[pyclass]
+pub struct PySwarmEvent(FromSwarm);
+
+#[gen_stub_pymethods]
+#[pymethods]
+impl PySwarmEvent {
+    // probably a better way to do this, but...
+    fn downcast_discovered(&self) -> Option<String> {
+        if let FromSwarm::Discovered(peer_id) = self.0 {
+            Some(peer_id.to_base58())
+        } else {
+            None
+        }
+    }
+    fn downcast_expired(&self) -> Option<String> {
+        if let FromSwarm::Expired(peer_id) = self.0 {
+            Some(peer_id.to_base58())
+        } else {
+            None
+        }
+    }
+    fn downcast_message<'py>(
+        &self,
+        py: Python<'py>,
+    ) -> Option<(String, String, Bound<'py, PyBytes>)> {
+        if let FromSwarm::Message(peer_id, topic, data) = &self.0 {
+            Some((peer_id.to_base58(), topic.clone(), PyBytes::new(py, data)))
+        } else {
+            None
+        }
+    }
+}

rust/exo_pyo3_bindings/tests/test_python.py

@@ -0,0 +1,34 @@
+import asyncio
+
+import pytest
+from exo_pyo3_bindings import Keypair, NetworkingHandle, NoPeersSubscribedToTopicError
+
+
+@pytest.mark.asyncio
+async def test_sleep_on_multiple_items() -> None:
+    print("PYTHON: starting handle")
+    h = NetworkingHandle(Keypair.generate_ed25519())
+
+    ct = asyncio.create_task(_await_cons(h))
+    mt = asyncio.create_task(_await_msg(h))
+
+    # sleep for 4 ticks
+    for i in range(4):
+        await asyncio.sleep(1)
+
+        try:
+            await h.gossipsub_publish("topic", b"somehting or other")
+        except NoPeersSubscribedToTopicError as e:
+            print("caught it", e)
+
+
+async def _await_cons(h: NetworkingHandle):
+    while True:
+        c = await h.connection_update_recv()
+        print(f"PYTHON: connection update: {c}")
+
+
+async def _await_msg(h: NetworkingHandle):
+    while True:
+        m = await h.gossipsub_recv()
+        print(f"PYTHON: message: {m}")

rust/networking/Cargo.toml

@@ -5,6 +5,7 @@ edition = { workspace = true }
 publish = false
 
 [lib]
+doctest = false
 name = "networking"
 path = "src/lib.rs"
 
@@ -12,16 +13,11 @@ path = "src/lib.rs"
 workspace = true
 
 [dependencies]
-# macro dependencies
-extend = { workspace = true }
-
 # async
 tokio = { workspace = true, features = ["full"] }
-futures-lite = { workspace = true }
 
 # utility dependencies
 tracing-subscriber = { version = "0.3.19", features = ["default", "env-filter"] }
-keccak-const = { workspace = true }
 
 # tracing/logging
 log = { workspace = true }

rust/networking/examples/chatroom.rs

@@ -1,3 +1,7 @@
+use libp2p::identity;
+use networking::{self, FromSwarm, ToSwarm};
+use tokio::sync::mpsc;
+use tokio::{io, io::AsyncBufReadExt as _, select};
 use tracing_subscriber::EnvFilter;
 use tracing_subscriber::filter::LevelFilter;
 
@@ -6,4 +10,53 @@ async fn main() {
     let _ = tracing_subscriber::fmt()
         .with_env_filter(EnvFilter::from_default_env().add_directive(LevelFilter::INFO.into()))
         .try_init();
+
+    // Configure swarm
+    let (to_client, mut from_swarm) = mpsc::channel(20);
+    let (to_swarm, from_client) = mpsc::channel(20);
+    let mut peer = networking::Peer::new(
+        "chatroom!".to_string(),
+        identity::Keypair::generate_ed25519(),
+        to_client,
+        from_client,
+    )
+    .expect("listen error");
+
+    // Create a Gossipsub topic & subscribe
+    // Read full lines from stdin
+    let mut stdin = io::BufReader::new(io::stdin()).lines();
+    println!("Enter messages via STDIN and they will be sent to connected peers using Gossipsub");
+
+    let jh = tokio::spawn(async move { peer.run().await });
+    _ = to_swarm
+        .send(ToSwarm::Subscribe("chatting".to_string()))
+        .await;
+
+    // Kick it off
+    loop {
+        select! {
+            // on gossipsub outgoing
+            Ok(Some(line)) = stdin.next_line() => {
+                _ = to_swarm.send(ToSwarm::Message("chatting".to_string(), line.into_bytes())).await;
+            }
+            event = from_swarm.recv() => match event {
+                // on gossipsub incoming
+                Some(FromSwarm::Message(peer_id,_, data)) => println!(
+                        "\n\nGot message: '{}' from peer: {peer_id}\n\n",
+                        String::from_utf8_lossy(&data),
+                    ),
+
+                // on discovery
+                Some(FromSwarm::Discovered(peer_id)) => {
+                    println!("\n\nConnected to: {peer_id}\n\n");
+                }
+                Some(FromSwarm::Expired(peer_id)) => {
+                    println!("\n\nDisconnected from: {peer_id}\n\n");
+                }
+                Some(FromSwarm::PublishError(e)) => eprintln!("\n\nError {e:?}\n\n"),
+                None => break,
+            }
+        }
+    }
+    _ = jh.await;
 }

rust/networking/src/lib.rs

@@ -0,0 +1,299 @@
+use libp2p::{
+    Multiaddr, PeerId,
+    futures::StreamExt,
+    gossipsub::{self, TopicHash},
+    identify,
+    identity::Keypair,
+    mdns,
+    swarm::{NetworkBehaviour, SwarmEvent, dial_opts::DialOpts},
+};
+use std::collections::HashMap;
+use tokio::sync::mpsc;
+
+#[derive(Debug)]
+pub struct ListenError;
+
+pub enum FromSwarm {
+    PublishError(gossipsub::PublishError),
+    Discovered(PeerId),
+    Expired(PeerId),
+    Message(PeerId, String, Vec<u8>),
+}
+pub enum ToSwarm {
+    Message(String, Vec<u8>),
+    Subscribe(String),
+    Unsubscribe(String),
+}
+
+pub struct Peer {
+    pub swarm: libp2p::Swarm<Behaviour>,
+    to_client: mpsc::Sender<FromSwarm>,
+    from_client: mpsc::Receiver<ToSwarm>,
+    namespace: String,
+    known_peers: HashMap<PeerId, Vec<Multiaddr>>,
+}
+impl Peer {
+    pub fn new(
+        namespace: String,
+        kp: Keypair,
+        to_client: mpsc::Sender<FromSwarm>,
+        from_client: mpsc::Receiver<ToSwarm>,
+    ) -> Result<Self, ListenError> {
+        let mut swarm = libp2p::SwarmBuilder::with_existing_identity(kp)
+            .with_tokio()
+            .with_quic()
+            // TODO(evan) .with_bandwidth_metrics()
+            .with_behaviour(|kp| Behaviour::new(namespace.clone(), kp))
+            .expect("invalid swarm behaviour")
+            .build();
+
+        swarm
+            .listen_on("/ip6/::/udp/0/quic-v1".parse().expect("invalid multiaddr"))
+            .map_err(|_| ListenError)?;
+        swarm
+            .listen_on(
+                "/ip4/0.0.0.0/udp/0/quic-v1"
+                    .parse()
+                    .expect("invalid multiaddr"),
+            )
+            .map_err(|_| ListenError)?;
+        Ok(Self {
+            swarm,
+            to_client,
+            from_client,
+            namespace,
+            known_peers: HashMap::default(),
+        })
+    }
+    pub async fn run(&mut self) -> Result<(), ()> {
+        loop {
+            tokio::select! {
+                event = self.swarm.next() => self.handle_event(event.ok_or(())?).await?,
+                msg = self.from_client.recv() => self.handle_message(msg.ok_or(())?).await?,
+            }
+        }
+    }
+    async fn handle_message(&mut self, message: ToSwarm) -> Result<(), ()> {
+        match message {
+            ToSwarm::Message(topic, data) => {
+                if let Err(e) = self
+                    .swarm
+                    .behaviour_mut()
+                    .gossipsub
+                    .publish(TopicHash::from_raw(topic), data)
+                {
+                    self.to_client
+                        .send(FromSwarm::PublishError(e))
+                        .await
+                        .map_err(|_| ())?;
+                }
+            }
+            ToSwarm::Subscribe(topic) => {
+                match self
+                    .swarm
+                    .behaviour_mut()
+                    .gossipsub
+                    .subscribe(&gossipsub::IdentTopic::new(topic))
+                {
+                    Ok(_) => {}
+                    Err(gossipsub::SubscriptionError::NotAllowed) => {
+                        unreachable!("subscription filter hit")
+                    }
+                    Err(gossipsub::SubscriptionError::PublishError(e)) => self
+                        .to_client
+                        .send(FromSwarm::PublishError(e))
+                        .await
+                        .map_err(|_| ())?,
+                }
+            }
+            ToSwarm::Unsubscribe(topic) => {
+                self.swarm
+                    .behaviour_mut()
+                    .gossipsub
+                    .unsubscribe(&gossipsub::IdentTopic::new(topic));
+            }
+        }
+        Ok(())
+    }
+    async fn handle_event(&mut self, event: SwarmEvent<BehaviourEvent>) -> Result<(), ()> {
+        let SwarmEvent::Behaviour(event) = event else {
+            return Ok(());
+        };
+        match event {
+            BehaviourEvent::Gossipsub(gossipsub::Event::Message { message, .. }) => {
+                if let Some(source) = message.source {
+                    self.to_client
+                        .send(FromSwarm::Message(
+                            source,
+                            message.topic.into_string(),
+                            message.data,
+                        ))
+                        .await
+                        .map_err(|_| ())?;
+                }
+            }
+            BehaviourEvent::Identify(identify::Event::Received { peer_id, info, .. }) => {
+                log::debug!(
+                    "identify from {peer_id}: protocol_version='{}' agent_version='{}' (local namespace='{}')",
+                    info.protocol_version,
+                    info.agent_version,
+                    self.namespace
+                );
+                if info.protocol_version == self.namespace {
+                    self.passed_namespace(peer_id);
+                    self.to_client
+                        .send(FromSwarm::Discovered(peer_id))
+                        .await
+                        .map_err(|_| ())?;
+                } else {
+                    self.failed_namespace(peer_id);
+                }
+            }
+            BehaviourEvent::Mdns(mdns::Event::Discovered(v)) => {
+                for (peer_id, addr) in v {
+                    self.known_peers.entry(peer_id).or_default().push(addr);
+                }
+                for (peer_id, addrs) in &self.known_peers {
+                    // dialopts handles rate limiting, we should check errors if we want to blacklist earlier
+                    let _ = self
+                        .swarm
+                        .dial(DialOpts::peer_id(*peer_id).addresses(addrs.clone()).build());
+                }
+            }
+            BehaviourEvent::Mdns(mdns::Event::Expired(v)) => {
+                for (peer_id, addr) in v {
+                    let addrs = self.known_peers.entry(peer_id).or_default();
+                    addrs.retain(|a| *a != addr);
+                    if addrs.is_empty() {
+                        self.known_peers.remove(&peer_id);
+                        self.swarm
+                            .behaviour_mut()
+                            .gossipsub
+                            .remove_explicit_peer(&peer_id);
+                        self.to_client
+                            .send(FromSwarm::Expired(peer_id))
+                            .await
+                            .map_err(|_| ())?;
+                    }
+                }
+            }
+            _ => {}
+        }
+        Ok(())
+    }
+    fn passed_namespace(&mut self, peer_id: PeerId) {
+        self.swarm
+            .behaviour_mut()
+            .gossipsub
+            .remove_blacklisted_peer(&peer_id);
+        self.swarm
+            .behaviour_mut()
+            .gossipsub
+            .add_explicit_peer(&peer_id);
+    }
+    fn failed_namespace(&mut self, peer_id: PeerId) {
+        self.swarm
+            .behaviour_mut()
+            .gossipsub
+            .blacklist_peer(&peer_id);
+        self.swarm
+            .behaviour_mut()
+            .gossipsub
+            .remove_explicit_peer(&peer_id);
+    }
+}
+
+#[derive(NetworkBehaviour)]
+pub struct Behaviour {
+    gossipsub: gossipsub::Behaviour,
+    mdns: mdns::tokio::Behaviour,
+    identify: identify::Behaviour,
+}
+
+impl Behaviour {
+    fn new(namespace: String, kp: &Keypair) -> Self {
+        let mdns = mdns::Behaviour::new(mdns::Config::default(), kp.public().to_peer_id())
+            .expect("mdns behaviour failed to build");
+
+        let identify =
+            identify::Behaviour::new(identify::Config::new_with_signed_peer_record(namespace, kp));
+
+        let gossipsub = gossipsub::Behaviour::new(
+            gossipsub::MessageAuthenticity::Signed(kp.clone()),
+            gossipsub::ConfigBuilder::default()
+                .max_transmit_size(1024 * 1024)
+                .validation_mode(gossipsub::ValidationMode::Strict)
+                .build()
+                .expect("invalid gossipsub configuration"),
+        )
+        .expect("gossipsub behaviour failed ot build");
+
+        Self {
+            gossipsub,
+            mdns,
+            identify,
+        }
+    }
+}
+
+// TODO: more tests
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use tokio::time::{Duration, timeout};
+
+    fn make_peer(namespace: &str) -> (Peer, mpsc::Receiver<FromSwarm>, mpsc::Sender<ToSwarm>) {
+        let kp = Keypair::generate_ed25519();
+
+        let (to_client_tx, to_client_rx) = mpsc::channel(64);
+        let (to_peer_tx, to_peer_rx) = mpsc::channel(64);
+
+        let peer = Peer::new(namespace.to_string(), kp, to_client_tx, to_peer_rx)
+            .expect("Peer::new should succeed in tests");
+
+        (peer, to_client_rx, to_peer_tx)
+    }
+
+    async fn next_listen_addr(peer: &mut Peer) -> Multiaddr {
+        loop {
+            match peer.swarm.next().await {
+                Some(SwarmEvent::NewListenAddr { address, .. }) => return address,
+                Some(_) => {}
+                None => panic!("swarm stream ended unexpectedly"),
+            }
+        }
+    }
+
+    #[tokio::test]
+    async fn subscribe_and_unsubscribe_do_not_error() {
+        let (mut peer, mut events_rx, commands_tx) = make_peer("ns-test");
+
+        // Drive the swarm just enough to get at least one listen address event,
+        // so the background run loop has something initialized.
+        let _addr = next_listen_addr(&mut peer).await;
+
+        // Run the peer loop in the background.
+        let handle = tokio::spawn(async move {
+            let _ = peer.run().await;
+        });
+
+        commands_tx
+            .send(ToSwarm::Subscribe("topic-a".to_string()))
+            .await
+            .unwrap();
+
+        commands_tx
+            .send(ToSwarm::Unsubscribe("topic-a".to_string()))
+            .await
+            .unwrap();
+
+        // We don't *require* any FromSwarm events here; this is mainly a
+        // smoke test that the message-handling path doesn't panic/hang.
+        // Still, poll briefly to ensure the task is alive.
+        let _ = timeout(Duration::from_millis(200), events_rx.recv()).await;
+
+        // Shut down: dropping the command sender closes the channel, causing run() to return Err.
+        drop(commands_tx);
+        let _ = handle.await;
+    }
+}

rust/rust-toolchain.toml

@@ -1,2 +0,0 @@
-[toolchain]
-channel = "nightly"

src/exo/download/coordinator.py

@@ -14,7 +14,6 @@ from exo.download.download_utils import (
     map_repo_download_progress_to_download_progress_data,
 )
 from exo.download.shard_downloader import ShardDownloader
-from exo.shared.constants import EXO_MODELS_DIR
 from exo.shared.models.model_cards import ModelId
 from exo.shared.types.commands import (
     CancelDownload,
@@ -47,7 +46,6 @@ class DownloadCoordinator:
     download_command_receiver: Receiver[ForwarderDownloadCommand]
     local_event_sender: Sender[ForwarderEvent]
     event_index_counter: Iterator[int]
-    offline: bool = False
 
     # Local state
     download_status: dict[ModelId, DownloadProgress] = field(default_factory=dict)
@@ -63,13 +61,8 @@ class DownloadCoordinator:
 
     def __post_init__(self) -> None:
         self.event_sender, self.event_receiver = channel[Event]()
-        if self.offline:
-            self.shard_downloader.set_internet_connection(False)
         self.shard_downloader.on_progress(self._download_progress_callback)
 
-    def _model_dir(self, model_id: ModelId) -> str:
-        return str(EXO_MODELS_DIR / model_id.normalize())
-
     async def _download_progress_callback(
         self, callback_shard: ShardMetadata, progress: RepoDownloadProgress
     ) -> None:
@@ -81,7 +74,6 @@ class DownloadCoordinator:
                 shard_metadata=callback_shard,
                 node_id=self.node_id,
                 total_bytes=progress.total_bytes,
-                model_directory=self._model_dir(model_id),
             )
             self.download_status[model_id] = completed
             await self.event_sender.send(
@@ -101,7 +93,6 @@ class DownloadCoordinator:
                 download_progress=map_repo_download_progress_to_download_progress_data(
                     progress
                 ),
-                model_directory=self._model_dir(model_id),
             )
             self.download_status[model_id] = ongoing
             await self.event_sender.send(
@@ -110,17 +101,13 @@ class DownloadCoordinator:
             self._last_progress_time[model_id] = current_time()
 
     async def run(self) -> None:
-        logger.info(
-            f"Starting DownloadCoordinator{' (offline mode)' if self.offline else ''}"
-        )
-        if not self.offline:
-            self._test_internet_connection()
+        logger.info("Starting DownloadCoordinator")
+        self._test_internet_connection()
         async with self._tg as tg:
             tg.start_soon(self._command_processor)
             tg.start_soon(self._forward_events)
             tg.start_soon(self._emit_existing_download_progress)
-            if not self.offline:
-                tg.start_soon(self._check_internet_connection)
+            tg.start_soon(self._check_internet_connection)
 
     def _test_internet_connection(self) -> None:
         try:
@@ -183,11 +170,7 @@ class DownloadCoordinator:
                 return
 
         # Emit pending status
-        progress = DownloadPending(
-            shard_metadata=shard,
-            node_id=self.node_id,
-            model_directory=self._model_dir(model_id),
-        )
+        progress = DownloadPending(shard_metadata=shard, node_id=self.node_id)
         self.download_status[model_id] = progress
         await self.event_sender.send(NodeDownloadProgress(download_progress=progress))
 
@@ -201,7 +184,6 @@ class DownloadCoordinator:
                 shard_metadata=shard,
                 node_id=self.node_id,
                 total_bytes=initial_progress.total_bytes,
-                model_directory=self._model_dir(model_id),
             )
             self.download_status[model_id] = completed
             await self.event_sender.send(
@@ -209,20 +191,6 @@ class DownloadCoordinator:
             )
             return
 
-        if self.offline:
-            logger.warning(
-                f"Offline mode: model {model_id} is not fully available locally, cannot download"
-            )
-            failed = DownloadFailed(
-                shard_metadata=shard,
-                node_id=self.node_id,
-                error_message=f"Model files not found locally in offline mode: {model_id}",
-                model_directory=self._model_dir(model_id),
-            )
-            self.download_status[model_id] = failed
-            await self.event_sender.send(NodeDownloadProgress(download_progress=failed))
-            return
-
         # Start actual download
         self._start_download_task(shard, initial_progress)
 
@@ -238,7 +206,6 @@ class DownloadCoordinator:
             download_progress=map_repo_download_progress_to_download_progress_data(
                 initial_progress
             ),
-            model_directory=self._model_dir(model_id),
         )
         self.download_status[model_id] = status
         self.event_sender.send_nowait(NodeDownloadProgress(download_progress=status))
@@ -252,7 +219,6 @@ class DownloadCoordinator:
                     shard_metadata=shard,
                     node_id=self.node_id,
                     error_message=str(e),
-                    model_directory=self._model_dir(model_id),
                 )
                 self.download_status[model_id] = failed
                 await self.event_sender.send(
@@ -287,7 +253,6 @@ class DownloadCoordinator:
             pending = DownloadPending(
                 shard_metadata=current_status.shard_metadata,
                 node_id=self.node_id,
-                model_directory=self._model_dir(model_id),
             )
             await self.event_sender.send(
                 NodeDownloadProgress(download_progress=pending)
@@ -330,18 +295,11 @@ class DownloadCoordinator:
                             node_id=self.node_id,
                             shard_metadata=progress.shard,
                             total_bytes=progress.total_bytes,
-                            model_directory=self._model_dir(
-                                progress.shard.model_card.model_id
-                            ),
                         )
                     elif progress.status in ["in_progress", "not_started"]:
                         if progress.downloaded_bytes_this_session.in_bytes == 0:
                             status = DownloadPending(
-                                node_id=self.node_id,
-                                shard_metadata=progress.shard,
-                                model_directory=self._model_dir(
-                                    progress.shard.model_card.model_id
-                                ),
+                                node_id=self.node_id, shard_metadata=progress.shard
                             )
                         else:
                             status = DownloadOngoing(
@@ -350,9 +308,6 @@ class DownloadCoordinator:
                                 download_progress=map_repo_download_progress_to_download_progress_data(
                                     progress
                                 ),
-                                model_directory=self._model_dir(
-                                    progress.shard.model_card.model_id
-                                ),
                             )
                     else:
                         continue

src/exo/download/download_utils.py

@@ -448,13 +448,12 @@ async def download_file_with_retry(
     target_dir: Path,
     on_progress: Callable[[int, int, bool], None] = lambda _, __, ___: None,
     on_connection_lost: Callable[[], None] = lambda: None,
-    skip_internet: bool = False,
 ) -> Path:
     n_attempts = 3
     for attempt in range(n_attempts):
         try:
             return await _download_file(
-                model_id, revision, path, target_dir, on_progress, skip_internet
+                model_id, revision, path, target_dir, on_progress
             )
         except HuggingFaceAuthenticationError:
             raise
@@ -488,14 +487,10 @@ async def _download_file(
     path: str,
     target_dir: Path,
     on_progress: Callable[[int, int, bool], None] = lambda _, __, ___: None,
-    skip_internet: bool = False,
 ) -> Path:
     target_path = target_dir / path
 
     if await aios.path.exists(target_path):
-        if skip_internet:
-            return target_path
-
         local_size = (await aios.stat(target_path)).st_size
 
         # Try to verify against remote, but allow offline operation
@@ -515,11 +510,6 @@ async def _download_file(
             )
             return target_path
 
-    if skip_internet:
-        raise FileNotFoundError(
-            f"File {path} not found locally and cannot download in offline mode"
-        )
-
     await aios.makedirs((target_dir / path).parent, exist_ok=True)
     length, etag = await file_meta(model_id, revision, path)
     remote_hash = etag[:-5] if etag.endswith("-gzip") else etag
@@ -824,7 +814,6 @@ async def download_shard(
                     file, curr_bytes, total_bytes, is_renamed
                 ),
                 on_connection_lost=on_connection_lost,
-                skip_internet=skip_internet,
             )
 
     if not skip_download:

src/exo/download/tests/test_offline_mode.py

@@ -1,230 +0,0 @@
-"""Tests for offline/air-gapped mode."""
-
-from collections.abc import AsyncIterator
-from pathlib import Path
-from unittest.mock import AsyncMock, patch
-
-import aiofiles
-import aiofiles.os as aios
-import pytest
-
-from exo.download.download_utils import (
-    _download_file,  # pyright: ignore[reportPrivateUsage]
-    download_file_with_retry,
-    fetch_file_list_with_cache,
-)
-from exo.shared.types.common import ModelId
-from exo.shared.types.worker.downloads import FileListEntry
-
-
-@pytest.fixture
-def model_id() -> ModelId:
-    return ModelId("test-org/test-model")
-
-
-@pytest.fixture
-async def temp_models_dir(tmp_path: Path) -> AsyncIterator[Path]:
-    models_dir = tmp_path / "models"
-    await aios.makedirs(models_dir, exist_ok=True)
-    with patch("exo.download.download_utils.EXO_MODELS_DIR", models_dir):
-        yield models_dir
-
-
-class TestDownloadFileOffline:
-    """Tests for _download_file with skip_internet=True."""
-
-    async def test_returns_local_file_without_http_verification(
-        self, model_id: ModelId, tmp_path: Path
-    ) -> None:
-        """When skip_internet=True and file exists locally, return it immediately
-        without making any HTTP calls (no file_meta verification)."""
-        target_dir = tmp_path / "downloads"
-        await aios.makedirs(target_dir, exist_ok=True)
-
-        local_file = target_dir / "model.safetensors"
-        async with aiofiles.open(local_file, "wb") as f:
-            await f.write(b"model weights data")
-
-        with patch(
-            "exo.download.download_utils.file_meta",
-            new_callable=AsyncMock,
-        ) as mock_file_meta:
-            result = await _download_file(
-                model_id,
-                "main",
-                "model.safetensors",
-                target_dir,
-                skip_internet=True,
-            )
-
-            assert result == local_file
-            mock_file_meta.assert_not_called()
-
-    async def test_raises_file_not_found_for_missing_file(
-        self, model_id: ModelId, tmp_path: Path
-    ) -> None:
-        """When skip_internet=True and file does NOT exist locally,
-        raise FileNotFoundError instead of attempting download."""
-        target_dir = tmp_path / "downloads"
-        await aios.makedirs(target_dir, exist_ok=True)
-
-        with pytest.raises(FileNotFoundError, match="offline mode"):
-            await _download_file(
-                model_id,
-                "main",
-                "missing_model.safetensors",
-                target_dir,
-                skip_internet=True,
-            )
-
-    async def test_returns_local_file_in_subdirectory(
-        self, model_id: ModelId, tmp_path: Path
-    ) -> None:
-        """When skip_internet=True and file exists in a subdirectory,
-        return it without HTTP calls."""
-        target_dir = tmp_path / "downloads"
-        subdir = target_dir / "transformer"
-        await aios.makedirs(subdir, exist_ok=True)
-
-        local_file = subdir / "diffusion_pytorch_model.safetensors"
-        async with aiofiles.open(local_file, "wb") as f:
-            await f.write(b"weights")
-
-        with patch(
-            "exo.download.download_utils.file_meta",
-            new_callable=AsyncMock,
-        ) as mock_file_meta:
-            result = await _download_file(
-                model_id,
-                "main",
-                "transformer/diffusion_pytorch_model.safetensors",
-                target_dir,
-                skip_internet=True,
-            )
-
-            assert result == local_file
-            mock_file_meta.assert_not_called()
-
-
-class TestDownloadFileWithRetryOffline:
-    """Tests for download_file_with_retry with skip_internet=True."""
-
-    async def test_propagates_skip_internet_to_download_file(
-        self, model_id: ModelId, tmp_path: Path
-    ) -> None:
-        """Verify skip_internet is passed through to _download_file."""
-        target_dir = tmp_path / "downloads"
-        await aios.makedirs(target_dir, exist_ok=True)
-
-        local_file = target_dir / "config.json"
-        async with aiofiles.open(local_file, "wb") as f:
-            await f.write(b'{"model_type": "qwen2"}')
-
-        with patch(
-            "exo.download.download_utils.file_meta",
-            new_callable=AsyncMock,
-        ) as mock_file_meta:
-            result = await download_file_with_retry(
-                model_id,
-                "main",
-                "config.json",
-                target_dir,
-                skip_internet=True,
-            )
-
-            assert result == local_file
-            mock_file_meta.assert_not_called()
-
-    async def test_file_not_found_does_not_retry(
-        self, model_id: ModelId, tmp_path: Path
-    ) -> None:
-        """FileNotFoundError from offline mode should not trigger retries."""
-        target_dir = tmp_path / "downloads"
-        await aios.makedirs(target_dir, exist_ok=True)
-
-        with pytest.raises(FileNotFoundError):
-            await download_file_with_retry(
-                model_id,
-                "main",
-                "nonexistent.safetensors",
-                target_dir,
-                skip_internet=True,
-            )
-
-
-class TestFetchFileListOffline:
-    """Tests for fetch_file_list_with_cache with skip_internet=True."""
-
-    async def test_uses_cached_file_list(
-        self, model_id: ModelId, temp_models_dir: Path
-    ) -> None:
-        """When skip_internet=True and cache file exists, use it without network."""
-        from pydantic import TypeAdapter
-
-        cache_dir = temp_models_dir / "caches" / model_id.normalize()
-        await aios.makedirs(cache_dir, exist_ok=True)
-
-        cached_list = [
-            FileListEntry(type="file", path="model.safetensors", size=1000),
-            FileListEntry(type="file", path="config.json", size=200),
-        ]
-        cache_file = cache_dir / f"{model_id.normalize()}--main--file_list.json"
-        async with aiofiles.open(cache_file, "w") as f:
-            await f.write(
-                TypeAdapter(list[FileListEntry]).dump_json(cached_list).decode()
-            )
-
-        with patch(
-            "exo.download.download_utils.fetch_file_list_with_retry",
-            new_callable=AsyncMock,
-        ) as mock_fetch:
-            result = await fetch_file_list_with_cache(
-                model_id, "main", skip_internet=True
-            )
-
-            assert result == cached_list
-            mock_fetch.assert_not_called()
-
-    async def test_falls_back_to_local_directory_scan(
-        self, model_id: ModelId, temp_models_dir: Path
-    ) -> None:
-        """When skip_internet=True and no cache but local files exist,
-        build file list from local directory."""
-        import json
-
-        model_dir = temp_models_dir / model_id.normalize()
-        await aios.makedirs(model_dir, exist_ok=True)
-
-        async with aiofiles.open(model_dir / "config.json", "w") as f:
-            await f.write('{"model_type": "qwen2"}')
-
-        index_data = {
-            "metadata": {},
-            "weight_map": {"model.layers.0.weight": "model.safetensors"},
-        }
-        async with aiofiles.open(model_dir / "model.safetensors.index.json", "w") as f:
-            await f.write(json.dumps(index_data))
-
-        async with aiofiles.open(model_dir / "model.safetensors", "wb") as f:
-            await f.write(b"x" * 500)
-
-        with patch(
-            "exo.download.download_utils.fetch_file_list_with_retry",
-            new_callable=AsyncMock,
-        ) as mock_fetch:
-            result = await fetch_file_list_with_cache(
-                model_id, "main", skip_internet=True
-            )
-
-            mock_fetch.assert_not_called()
-            paths = {entry.path for entry in result}
-            assert "config.json" in paths
-            assert "model.safetensors" in paths
-
-    async def test_raises_when_no_cache_and_no_local_files(
-        self, model_id: ModelId, temp_models_dir: Path
-    ) -> None:
-        """When skip_internet=True and neither cache nor local files exist,
-        raise FileNotFoundError."""
-        with pytest.raises(FileNotFoundError, match="No internet"):
-            await fetch_file_list_with_cache(model_id, "main", skip_internet=True)

src/exo/main.py

@@ -1,4 +1,5 @@
 import argparse
+import importlib.metadata
 import itertools
 import multiprocessing as mp
 import os
@@ -39,15 +40,14 @@ class Node:
 
     node_id: NodeId
     event_index_counter: Iterator[int]
-    offline: bool
     _tg: TaskGroup = field(init=False, default_factory=anyio.create_task_group)
 
     @classmethod
     async def create(cls, args: "Args") -> "Self":
         keypair = get_node_id_keypair()
-        node_id = NodeId(keypair.to_peer_id().to_base58())
+        node_id = NodeId(keypair.to_string())
         session_id = SessionId(master_node_id=node_id, election_clock=0)
-        router = Router.create(keypair)
+        router = Router.create(keypair, namespace=args.namespace)
         await router.register_topic(topics.GLOBAL_EVENTS)
         await router.register_topic(topics.LOCAL_EVENTS)
         await router.register_topic(topics.COMMANDS)
@@ -69,12 +69,11 @@ class Node:
                 download_command_receiver=router.receiver(topics.DOWNLOAD_COMMANDS),
                 local_event_sender=router.sender(topics.LOCAL_EVENTS),
                 event_index_counter=event_index_counter,
-                offline=args.offline,
             )
         else:
             download_coordinator = None
 
-        if args.spawn_api:
+        if not args.no_api:
             api = API(
                 node_id,
                 session_id,
@@ -134,13 +133,10 @@ class Node:
             api,
             node_id,
             event_index_counter,
-            args.offline,
         )
 
     async def run(self):
         async with self._tg as tg:
-            signal.signal(signal.SIGINT, lambda _, __: self.shutdown())
-            signal.signal(signal.SIGTERM, lambda _, __: self.shutdown())
             tg.start_soon(self.router.run)
             tg.start_soon(self.election.run)
             if self.download_coordinator:
@@ -152,6 +148,8 @@ class Node:
             if self.api:
                 tg.start_soon(self.api.run)
             tg.start_soon(self._elect_loop)
+            signal.signal(signal.SIGINT, lambda _, __: self.shutdown())
+            signal.signal(signal.SIGTERM, lambda _, __: self.shutdown())
 
     def shutdown(self):
         # if this is our second call to shutdown, just sys.exit
@@ -225,7 +223,6 @@ class Node:
                             ),
                             local_event_sender=self.router.sender(topics.LOCAL_EVENTS),
                             event_index_counter=self.event_index_counter,
-                            offline=self.offline,
                         )
                         self._tg.start_soon(self.download_coordinator.run)
                     if self.worker:
@@ -262,10 +259,7 @@ def main():
     # TODO: Refactor the current verbosity system
     logger_setup(EXO_LOG, args.verbosity)
     logger.info("Starting EXO")
-    logger.info(f"EXO_LIBP2P_NAMESPACE: {os.getenv('EXO_LIBP2P_NAMESPACE')}")
-
-    if args.offline:
-        logger.info("Running in OFFLINE mode — no internet checks, local models only")
+    logger.info(f"Namespace: {args.namespace}")
 
     # Set FAST_SYNCH override env var for runner subprocesses
     if args.fast_synch is True:
@@ -282,14 +276,13 @@ def main():
 
 
 class Args(CamelCaseModel):
-    verbosity: int = 0
-    force_master: bool = False
-    spawn_api: bool = False
-    api_port: PositiveInt = 52415
-    tb_only: bool = False
+    verbosity: int
+    force_master: bool
+    no_api: bool
+    api_port: PositiveInt
     no_worker: bool = False
     no_downloads: bool = False
-    offline: bool = False
+    namespace: str
     fast_synch: bool | None = None  # None = auto, True = force on, False = force off
 
     @classmethod
@@ -319,14 +312,15 @@ class Args(CamelCaseModel):
         )
         parser.add_argument(
             "--no-api",
-            action="store_false",
-            dest="spawn_api",
+            action="store_true",
+            help="Disable the API server for this node",
         )
         parser.add_argument(
             "--api-port",
             type=int,
             dest="api_port",
             default=52415,
+            help="Which port the API server will be available on",
         )
         parser.add_argument(
             "--no-worker",
@@ -338,9 +332,9 @@ class Args(CamelCaseModel):
             help="Disable the download coordinator (node won't download models)",
         )
         parser.add_argument(
-            "--offline",
-            action="store_true",
-            help="Run in offline/air-gapped mode: skip internet checks, use only pre-staged local models",
+            "--namespace",
+            default=importlib.metadata.version("exo"),
+            help="Set the EXO namespace to run multiple isolated clusters",
         )
         fast_synch_group = parser.add_mutually_exclusive_group()
         fast_synch_group.add_argument(

src/exo/master/api.py

@@ -85,7 +85,6 @@ from exo.shared.types.api import (
     ImageGenerationTaskParams,
     ImageListItem,
     ImageListResponse,
-    ImageSize,
     ModelList,
     ModelListModel,
     PlaceInstanceParams,
@@ -101,7 +100,6 @@ from exo.shared.types.api import (
     TraceRankStats,
     TraceResponse,
     TraceStatsResponse,
-    normalize_image_size,
 )
 from exo.shared.types.chunks import (
     ErrorChunk,
@@ -753,11 +751,9 @@ class API:
         When stream=True and partial_images > 0, returns a StreamingResponse
         with SSE-formatted events for partial and final images.
         """
+        payload.model = await self._validate_image_model(ModelId(payload.model))
         payload = payload.model_copy(
-            update={
-                "model": await self._validate_image_model(ModelId(payload.model)),
-                "advanced_params": _ensure_seed(payload.advanced_params),
-            }
+            update={"advanced_params": _ensure_seed(payload.advanced_params)}
         )
 
         command = ImageGeneration(
@@ -1013,13 +1009,12 @@ class API:
     async def bench_image_generations(
         self, request: Request, payload: BenchImageGenerationTaskParams
     ) -> BenchImageGenerationResponse:
+        payload.model = await self._validate_image_model(ModelId(payload.model))
+
+        payload.stream = False
+        payload.partial_images = 0
         payload = payload.model_copy(
-            update={
-                "model": await self._validate_image_model(ModelId(payload.model)),
-                "stream": False,
-                "partial_images": 0,
-                "advanced_params": _ensure_seed(payload.advanced_params),
-            }
+            update={"advanced_params": _ensure_seed(payload.advanced_params)}
         )
 
         command = ImageGeneration(
@@ -1040,7 +1035,7 @@ class API:
         prompt: str,
         model: ModelId,
         n: int,
-        size: ImageSize,
+        size: str,
         response_format: Literal["url", "b64_json"],
         input_fidelity: Literal["low", "high"],
         stream: bool,
@@ -1110,7 +1105,7 @@ class API:
         prompt: str = Form(...),
         model: str = Form(...),
         n: int = Form(1),
-        size: str | None = Form(None),
+        size: str = Form("1024x1024"),
         response_format: Literal["url", "b64_json"] = Form("b64_json"),
         input_fidelity: Literal["low", "high"] = Form("low"),
         stream: str = Form("false"),
@@ -1136,7 +1131,7 @@ class API:
             prompt=prompt,
             model=ModelId(model),
             n=n,
-            size=normalize_image_size(size),
+            size=size,
             response_format=response_format,
             input_fidelity=input_fidelity,
             stream=stream_bool,
@@ -1172,7 +1167,7 @@ class API:
         prompt: str = Form(...),
         model: str = Form(...),
         n: int = Form(1),
-        size: str | None = Form(None),
+        size: str = Form("1024x1024"),
         response_format: Literal["url", "b64_json"] = Form("b64_json"),
         input_fidelity: Literal["low", "high"] = Form("low"),
         quality: Literal["high", "medium", "low"] = Form("medium"),
@@ -1192,7 +1187,7 @@ class API:
             prompt=prompt,
             model=ModelId(model),
             n=n,
-            size=normalize_image_size(size),
+            size=size,
             response_format=response_format,
             input_fidelity=input_fidelity,
             stream=False,

src/exo/master/main.py

@@ -396,7 +396,7 @@ class Master:
                         await self._handle_traces_collected(event)
                         continue
 
-                    logger.debug(f"Master indexing event: {str(event)[:100]}")
+                    logger.trace(f"Master indexing event: {str(event)[:100]}")
                     indexed = IndexedEvent(event=event, idx=len(self._event_log))
                     self.state = apply(self.state, indexed)
 

src/exo/master/tests/test_master.py

@@ -42,7 +42,7 @@ from exo.utils.channels import channel
 @pytest.mark.asyncio
 async def test_master():
     keypair = get_node_id_keypair()
-    node_id = NodeId(keypair.to_peer_id().to_base58())
+    node_id = NodeId(keypair.to_string())
     session_id = SessionId(master_node_id=node_id, election_clock=0)
 
     ge_sender, global_event_receiver = channel[ForwarderEvent]()
@@ -75,7 +75,7 @@ async def test_master():
     async with anyio.create_task_group() as tg:
         tg.start_soon(master.run)
 
-        sender_node_id = NodeId(f"{keypair.to_peer_id().to_base58()}_sender")
+        sender_node_id = NodeId(f"{keypair.to_string()}_sender")
         # inject a NodeGatheredInfo event
         logger.info("inject a NodeGatheredInfo event")
         await local_event_sender.send(

src/exo/routing/connection_message.py

@@ -1,37 +1,9 @@
-from enum import Enum
-
-from exo_pyo3_bindings import ConnectionUpdate, ConnectionUpdateType
-
 from exo.shared.types.common import NodeId
 from exo.utils.pydantic_ext import CamelCaseModel
 
 """Serialisable types for Connection Updates/Messages"""
 
 
-class ConnectionMessageType(Enum):
-    Connected = 0
-    Disconnected = 1
-
-    @staticmethod
-    def from_update_type(update_type: ConnectionUpdateType):
-        match update_type:
-            case ConnectionUpdateType.Connected:
-                return ConnectionMessageType.Connected
-            case ConnectionUpdateType.Disconnected:
-                return ConnectionMessageType.Disconnected
-
-
 class ConnectionMessage(CamelCaseModel):
     node_id: NodeId
-    connection_type: ConnectionMessageType
-    remote_ipv4: str
-    remote_tcp_port: int
-
-    @classmethod
-    def from_update(cls, update: ConnectionUpdate) -> "ConnectionMessage":
-        return cls(
-            node_id=NodeId(update.peer_id.to_base58()),
-            connection_type=ConnectionMessageType.from_update_type(update.update_type),
-            remote_ipv4=update.remote_ipv4,
-            remote_tcp_port=update.remote_tcp_port,
-        )
+    expired: bool

src/exo/routing/router.py

@@ -1,5 +1,5 @@
 from copy import copy
-from itertools import count
+from dataclasses import dataclass, field
 from math import inf
 from os import PathLike
 from pathlib import Path
@@ -14,15 +14,14 @@ from anyio import (
 )
 from anyio.abc import TaskGroup
 from exo_pyo3_bindings import (
-    AllQueuesFullError,
     Keypair,
-    NetworkingHandle,
-    NoPeersSubscribedToTopicError,
+    PyPeer,
 )
 from filelock import FileLock
 from loguru import logger
 
 from exo.shared.constants import EXO_NODE_ID_KEYPAIR
+from exo.shared.types.common import NodeId
 from exo.utils.channels import Receiver, Sender, channel
 from exo.utils.pydantic_ext import CamelCaseModel
 
@@ -99,28 +98,32 @@ class TopicRouter[T: CamelCaseModel]:
         )
 
 
+@dataclass
 class Router:
-    @classmethod
-    def create(cls, identity: Keypair) -> "Router":
-        return cls(handle=NetworkingHandle(identity))
+    _peer: PyPeer
+    topic_routers: dict[str, TopicRouter[CamelCaseModel]] = field(
+        init=False, default_factory=dict
+    )
+    networking_receiver: Receiver[tuple[str, bytes]] = field(init=False)
+    _tmp_networking_sender: Sender[tuple[str, bytes]] | None = field(init=False)
+    _tg: TaskGroup | None = None
 
-    def __init__(self, handle: NetworkingHandle):
-        self.topic_routers: dict[str, TopicRouter[CamelCaseModel]] = {}
-        send, recv = channel[tuple[str, bytes]]()
-        self.networking_receiver: Receiver[tuple[str, bytes]] = recv
-        self._net: NetworkingHandle = handle
-        self._tmp_networking_sender: Sender[tuple[str, bytes]] | None = send
-        self._id_count = count()
-        self._tg: TaskGroup | None = None
+    def __post_init__(self):
+        self._tmp_networking_sender, self.networking_receiver = channel()
+
+    @classmethod
+    def create(cls, identity: Keypair, namespace: str) -> "Router":
+        return cls(_peer=PyPeer.new(identity, namespace))
 
     async def register_topic[T: CamelCaseModel](self, topic: TypedTopic[T]):
-        assert self._tg is None, "Attempted to register topic after setup time"
         send = self._tmp_networking_sender
         if send:
             self._tmp_networking_sender = None
         else:
             send = self.networking_receiver.clone_sender()
         router = TopicRouter[T](topic, send)
+        if self._tg is not None:
+            self._tg.start_soon(router.run)
         self.topic_routers[topic.topic] = cast(TopicRouter[CamelCaseModel], router)
         await self._networking_subscribe(str(topic.topic))
 
@@ -148,14 +151,18 @@ class Router:
     async def run(self):
         logger.debug("Starting Router")
         try:
+
+            async def _peer_run():
+                await self._peer.run()
+
             async with create_task_group() as tg:
                 self._tg = tg
                 for topic in self.topic_routers:
                     router = self.topic_routers[topic]
                     tg.start_soon(router.run)
                 tg.start_soon(self._networking_recv)
-                tg.start_soon(self._networking_recv_connection_messages)
                 tg.start_soon(self._networking_publish)
+                tg.start_soon(_peer_run)
                 # Router only shuts down if you cancel it.
                 await sleep_forever()
         finally:
@@ -170,47 +177,58 @@ class Router:
         self._tg.cancel_scope.cancel()
 
     async def _networking_subscribe(self, topic: str):
-        await self._net.gossipsub_subscribe(topic)
+        await self._peer.subscribe(topic)
         logger.info(f"Subscribed to {topic}")
 
     async def _networking_unsubscribe(self, topic: str):
-        await self._net.gossipsub_unsubscribe(topic)
+        await self._peer.unsubscribe(topic)
         logger.info(f"Unsubscribed from {topic}")
 
     async def _networking_recv(self):
         while True:
-            topic, data = await self._net.gossipsub_recv()
-            logger.trace(f"Received message on {topic} with payload {data}")
+            try:
+                swarm_event = await self._peer.recv()
+            except ValueError:
+                logger.error("Message too large for gossipsub, dropped")
+                continue
+            except ConnectionError:
+                logger.error("All peer queues full, network overloaded")
+                continue
+            except RuntimeError:
+                break
+
+            cm = None
+            if (peer_id := swarm_event.downcast_discovered()) is not None:
+                cm = ConnectionMessage(node_id=NodeId(peer_id), expired=False)
+            if (peer_id := swarm_event.downcast_expired()) is not None:
+                cm = ConnectionMessage(node_id=NodeId(peer_id), expired=True)
+
+            if cm is not None:
+                if CONNECTION_MESSAGES.topic in self.topic_routers:
+                    router = self.topic_routers[CONNECTION_MESSAGES.topic]
+                    assert router.topic.model_type == ConnectionMessage
+                    router = cast(TopicRouter[ConnectionMessage], router)
+                    await router.publish(cm)
+                continue
+
+            assert (msg := swarm_event.downcast_message()) is not None
+            _origin, topic, payload = msg
+            logger.debug(f"Received message on {topic} with payload {payload}")
             if topic not in self.topic_routers:
                 logger.warning(f"Received message on unknown or inactive topic {topic}")
                 continue
 
             router = self.topic_routers[topic]
-            await router.publish_bytes(data)
-
-    async def _networking_recv_connection_messages(self):
-        while True:
-            update = await self._net.connection_update_recv()
-            message = ConnectionMessage.from_update(update)
-            logger.trace(
-                f"Received message on connection_messages with payload {message}"
-            )
-            if CONNECTION_MESSAGES.topic in self.topic_routers:
-                router = self.topic_routers[CONNECTION_MESSAGES.topic]
-                assert router.topic.model_type == ConnectionMessage
-                router = cast(TopicRouter[ConnectionMessage], router)
-                await router.publish(message)
+            await router.publish_bytes(payload)
 
     async def _networking_publish(self):
         with self.networking_receiver as networked_items:
             async for topic, data in networked_items:
                 try:
                     logger.trace(f"Sending message on {topic} with payload {data}")
-                    await self._net.gossipsub_publish(topic, data)
-                except NoPeersSubscribedToTopicError:
-                    pass
-                except AllQueuesFullError:
-                    logger.warning(f"All peer queues full, dropping message on {topic}")
+                    await self._peer.send(topic, data)
+                except RuntimeError:
+                    break
 
 
 def get_node_id_keypair(
@@ -221,7 +239,7 @@ def get_node_id_keypair(
     Obtain the :class:`PeerId` by from it.
     """
     # TODO(evan): bring back node id persistence once we figure out how to deal with duplicates
-    return Keypair.generate_ed25519()
+    return Keypair.generate()
 
     def lock_path(path: str | bytes | PathLike[str] | PathLike[bytes]) -> Path:
         return Path(str(path) + ".lock")

src/exo/shared/apply.py

@@ -218,6 +218,11 @@ def apply_node_timed_out(event: NodeTimedOut, state: State) -> State:
         key: value for key, value in state.downloads.items() if key != event.node_id
     }
     # Clean up all granular node mappings
+    node_identities = {
+        key: value
+        for key, value in state.node_identities.items()
+        if key != event.node_id
+    }
     node_memory = {
         key: value for key, value in state.node_memory.items() if key != event.node_id
     }
@@ -258,6 +263,7 @@ def apply_node_timed_out(event: NodeTimedOut, state: State) -> State:
             "downloads": downloads,
             "topology": topology,
             "last_seen": last_seen,
+            "node_identities": node_identities,
             "node_memory": node_memory,
             "node_disk": node_disk,
             "node_system": node_system,

src/exo/shared/models/model_cards.py

@@ -44,8 +44,7 @@ async def _refresh_card_cache():
         async for toml_file in path.rglob("*.toml"):
             try:
                 card = await ModelCard.load_from_path(toml_file)
-                if card.model_id not in _card_cache:
-                    _card_cache[card.model_id] = card
+                _card_cache[card.model_id] = card
             except (ValidationError, TOMLKitError):
                 pass
 
@@ -183,7 +182,6 @@ class ConfigData(BaseModel):
     def supports_tensor(self) -> bool:
         return self.architectures in [
             ["Glm4MoeLiteForCausalLM"],
-            ["GlmMoeDsaForCausalLM"],
             ["DeepseekV32ForCausalLM"],
             ["DeepseekV3ForCausalLM"],
             ["Qwen3NextForCausalLM"],

src/exo/shared/tests/test_election.py

@@ -1,7 +1,7 @@
 import pytest
 from anyio import create_task_group, fail_after, move_on_after
 
-from exo.routing.connection_message import ConnectionMessage, ConnectionMessageType
+from exo.routing.connection_message import ConnectionMessage
 from exo.shared.election import Election, ElectionMessage, ElectionResult
 from exo.shared.types.commands import ForwarderCommand, TestCommand
 from exo.shared.types.common import NodeId, SessionId
@@ -330,9 +330,7 @@ async def test_connection_message_triggers_new_round_broadcast() -> None:
             await cm_tx.send(
                 ConnectionMessage(
                     node_id=NodeId(),
-                    connection_type=ConnectionMessageType.Connected,
-                    remote_ipv4="",
-                    remote_tcp_port=0,
+                    expired=False,
                 )
             )
 

src/exo/shared/types/api.py

@@ -1,9 +1,9 @@
 import time
 from collections.abc import Generator
-from typing import Annotated, Any, Literal, get_args
+from typing import Annotated, Any, Literal
 from uuid import uuid4
 
-from pydantic import BaseModel, Field, field_validator
+from pydantic import BaseModel, Field
 
 from exo.shared.models.model_cards import ModelCard, ModelId
 from exo.shared.types.common import CommandId, NodeId
@@ -262,27 +262,6 @@ class DeleteInstanceResponse(BaseModel):
     instance_id: InstanceId
 
 
-ImageSize = Literal[
-    "auto",
-    "512x512",
-    "768x768",
-    "1024x768",
-    "768x1024",
-    "1024x1024",
-    "1024x1536",
-    "1536x1024",
-]
-
-
-def normalize_image_size(v: object) -> ImageSize:
-    """Shared validator for ImageSize fields: maps None → "auto" and rejects invalid values."""
-    if v is None:
-        return "auto"
-    if v not in get_args(ImageSize):
-        raise ValueError(f"Invalid size: {v!r}. Must be one of {get_args(ImageSize)}")
-    return v  # pyright: ignore[reportReturnType]
-
-
 class AdvancedImageParams(BaseModel):
     seed: Annotated[int, Field(ge=0)] | None = None
     num_inference_steps: Annotated[int, Field(ge=1, le=100)] | None = None
@@ -302,7 +281,7 @@ class ImageGenerationTaskParams(BaseModel):
     partial_images: int | None = 0
     quality: Literal["high", "medium", "low"] | None = "medium"
     response_format: Literal["url", "b64_json"] | None = "b64_json"
-    size: ImageSize = "auto"
+    size: str | None = "1024x1024"
     stream: bool | None = False
     style: str | None = "vivid"
     user: str | None = None
@@ -310,11 +289,6 @@ class ImageGenerationTaskParams(BaseModel):
     # Internal flag for benchmark mode - set by API, preserved through serialization
     bench: bool = False
 
-    @field_validator("size", mode="before")
-    @classmethod
-    def normalize_size(cls, v: object) -> ImageSize:
-        return normalize_image_size(v)
-
 
 class BenchImageGenerationTaskParams(ImageGenerationTaskParams):
     bench: bool = True
@@ -331,18 +305,13 @@ class ImageEditsTaskParams(BaseModel):
     quality: Literal["high", "medium", "low"] | None = "medium"
     output_format: Literal["png", "jpeg", "webp"] = "png"
     response_format: Literal["url", "b64_json"] | None = "b64_json"
-    size: ImageSize = "auto"
+    size: str | None = "1024x1024"
     image_strength: float | None = 0.7
     stream: bool = False
     partial_images: int | None = 0
     advanced_params: AdvancedImageParams | None = None
     bench: bool = False
 
-    @field_validator("size", mode="before")
-    @classmethod
-    def normalize_size(cls, v: object) -> ImageSize:
-        return normalize_image_size(v)
-
     def __repr_args__(self) -> Generator[tuple[str, Any], None, None]:
         for name, value in super().__repr_args__():  # pyright: ignore[reportAny]
             if name == "image_data":

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

@@ -26,7 +26,6 @@ class DownloadProgressData(CamelCaseModel):
 class BaseDownloadProgress(TaggedModel):
     node_id: NodeId
     shard_metadata: ShardMetadata
-    model_directory: str = ""
 
 
 class DownloadPending(BaseDownloadProgress):

src/exo/worker/engines/image/generate.py

@@ -14,7 +14,6 @@ from exo.shared.types.api import (
     ImageEditsTaskParams,
     ImageGenerationStats,
     ImageGenerationTaskParams,
-    ImageSize,
 )
 from exo.shared.types.memory import Memory
 from exo.shared.types.worker.runner_response import (
@@ -24,9 +23,9 @@ from exo.shared.types.worker.runner_response import (
 from exo.worker.engines.image.distributed_model import DistributedImageModel
 
 
-def parse_size(size_str: ImageSize) -> tuple[int, int]:
+def parse_size(size_str: str | None) -> tuple[int, int]:
     """Parse size parameter like '1024x1024' to (width, height) tuple."""
-    if size_str == "auto":
+    if not size_str:
         return (1024, 1024)
 
     try:
@@ -110,9 +109,6 @@ def generate_image(
             # Decode base64 image data and save to temp file
             image_path = Path(tmpdir) / "input.png"
             image_path.write_bytes(base64.b64decode(task.image_data))
-            if task.size == "auto":
-                with Image.open(image_path) as img:
-                    width, height = img.size
 
         for image_num in range(num_images):
             # Increment seed for each image to ensure unique results

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

@@ -163,14 +163,11 @@ class PipelineLastLayer(CustomMlxLayer):
                 output, (self.r + 1) % self.s, group=self.group
             )
             if cache is not None:
-                # CacheList (used by MLA models like DeepSeekV32, GLM MoE DSA)
-                # doesn't have .keys directly; access via first sub-cache.
-                _cache = cache[0] if hasattr(cache, "caches") else cache  # type: ignore
-                _cache.keys = mx.depends(_cache.keys, output)  # type: ignore
+                cache.keys = mx.depends(cache.keys, output)  # type: ignore[reportUnknownMemberType]
             if self.is_prefill:
                 mx.eval(output)
                 if cache is not None:
-                    mx.eval(_cache.keys)  # type: ignore
+                    mx.eval(cache.keys)  # type: ignore
 
         if not self.is_prefill:
             output = mx.distributed.all_gather(output, group=self.group)[
@@ -310,9 +307,7 @@ def patch_pipeline_model[T](model: T, group: mx.distributed.Group) -> T:
 
         # Add dependency to last cache entry to ensure distributed ops are evaluated
         if cache is not None:
-            last = cache[-1]  # type: ignore
-            dep_cache = last[0] if hasattr(last, "caches") else last  # type: ignore
-            dep_cache.keys = mx.depends(dep_cache.keys, logits)  # type: ignore
+            cache[-1].state = mx.depends(cache[-1].state, logits)  # type: ignore
 
         return logits
 
@@ -338,9 +333,7 @@ def patch_tensor_model[T](model: T) -> T:
 
         # Add dependency to last cache entry to ensure distributed ops are evaluated
         if cache is not None and len(cache) > 0:  # pyright: ignore[reportAny]
-            last = cache[-1]  # pyright: ignore[reportAny]
-            dep_cache = last[0] if hasattr(last, "caches") else last  # pyright: ignore[reportAny]
-            dep_cache.keys = mx.depends(dep_cache.keys, logits)  # pyright: ignore[reportAny,reportUnknownMemberType]
+            cache[-1].state = mx.depends(cache[-1].state, logits)  # pyright: ignore[reportAny,reportUnknownMemberType]
 
         return logits
 
@@ -554,12 +547,10 @@ class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
         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:
                 layer.self_attn.q_proj = self.all_to_sharded_linear(
@@ -590,18 +581,12 @@ class DeepSeekShardingStrategy(TensorParallelShardingStrategy):
                 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)
 
-            # Shard the MoE.
+            # Shard the MoE. Shard in place since the MoE should be responsible
+            # for aggregating the results.
             else:
-                if getattr(layer.mlp, "shared_experts", None) is not None:
-                    self.all_to_sharded_linear_in_place(
-                        layer.mlp.shared_experts.gate_proj
-                    )
-                    self.sharded_to_all_linear_in_place(
-                        layer.mlp.shared_experts.down_proj
-                    )
-                    self.all_to_sharded_linear_in_place(
-                        layer.mlp.shared_experts.up_proj
-                    )
+                self.all_to_sharded_linear_in_place(layer.mlp.shared_experts.gate_proj)
+                self.sharded_to_all_linear_in_place(layer.mlp.shared_experts.down_proj)
+                self.all_to_sharded_linear_in_place(layer.mlp.shared_experts.up_proj)
                 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)
@@ -794,7 +779,8 @@ class MiniMaxShardingStrategy(TensorParallelShardingStrategy):
 
             layer.self_attn = WrappedMiniMaxAttention(layer.self_attn, self.group)  # pyright: ignore[reportAttributeAccessIssue,reportArgumentType]
 
-            # Shard the MoE.
+            # 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
             )
@@ -907,7 +893,8 @@ class QwenShardingStrategy(TensorParallelShardingStrategy):
                     layer.self_attn.num_attention_heads //= self.N
                     layer.self_attn.num_key_value_heads //= self.N
 
-            # Shard the MoE.
+            # Shard the MoE. Shard in place since the MoE should be responsible
+            # for aggregating the results.
             if isinstance(layer.mlp, (Qwen3MoeSparseMoeBlock, 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)

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

@@ -57,7 +57,6 @@ def prefill(
     sampler: Callable[[mx.array], mx.array],
     prompt_tokens: mx.array,
     cache: KVCacheType,
-    group: mx.distributed.Group | None = None,
 ) -> tuple[float, int, list[CacheSnapshot]]:
     """Prefill the KV cache with prompt tokens.
 
@@ -87,9 +86,6 @@ def prefill(
 
     set_pipeline_prefill(model, is_prefill=True)
 
-    mx_barrier(group)
-    logger.info("Starting prefill")
-
     # Use max_tokens=1 because max_tokens=0 does not work.
     # We just throw away the generated token - we only care about filling the cache
     for _ in stream_generate(
@@ -309,9 +305,16 @@ def mlx_generate(
     )
     max_stop_len = max((len(s) for s in stop_sequences), default=0)
 
+    mx_barrier(group)
+    logger.info("Starting prefill")
+
     # Prefill cache with all tokens except the last one
     prefill_tps, prefill_tokens, ssm_snapshots_list = prefill(
-        model, tokenizer, sampler, prompt_tokens[:-1], caches, group
+        model,
+        tokenizer,
+        sampler,
+        prompt_tokens[:-1],
+        caches,
     )
     cache_snapshots: list[CacheSnapshot] | None = ssm_snapshots_list or None
 
@@ -328,7 +331,6 @@ def mlx_generate(
     think_start = tokenizer.think_start
     think_end = tokenizer.think_end
 
-    logger.info("Starting decode")
     mx_barrier(group)
 
     for completion_tokens, out in enumerate(

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

@@ -285,12 +285,10 @@ def get_eos_token_ids_for_model(model_id: ModelId) -> list[int] | None:
     model_id_lower = model_id.lower()
     if "kimi-k2" in model_id_lower:
         return [163586]
-    elif "glm-5" in model_id_lower or "glm-4.7" in model_id_lower:
-        # For GLM-5 and GLM-4.7
+    elif "glm-4.7-flash" in model_id_lower:
         # 154820: <|endoftext|>, 154827: <|user|>, 154829: <|observation|>
         return [154820, 154827, 154829]
     elif "glm" in model_id_lower:
-        # For GLM-4.5 and older
         return [151336, 151329, 151338]
     return None
 

src/exo/worker/main.py

@@ -348,7 +348,7 @@ class Worker:
                     session=self.session_id,
                     event=event,
                 )
-                logger.debug(f"Worker published event {idx}: {str(event)[:100]}")
+                logger.trace(f"Worker published event {idx}: {str(event)[:100]}")
                 await self.local_event_sender.send(fe)
                 self.out_for_delivery[event.event_id] = fe
 

src/exo/worker/runner/runner_supervisor.py

@@ -191,7 +191,7 @@ class RunnerSupervisor:
         logger.info("Checking runner's status")
         if self.runner_process.is_alive():
             logger.info("Runner was found to be alive, attempting to join process")
-            await to_thread.run_sync(self.runner_process.join, 5)
+            await to_thread.run_sync(self.runner_process.join, 1)
         rc = self.runner_process.exitcode
         logger.info(f"RunnerSupervisor exited with exit code {rc}")
         if rc == 0:

tests/auto_bench.sh

@@ -28,12 +28,12 @@ trap 'cleanup' EXIT INT TERM
 
 for host; do
   ssh -T -o BatchMode=yes -o ServerAliveInterval=30 "$host@$host" \
-    "EXO_LIBP2P_NAMESPACE=$commit /nix/var/nix/profiles/default/bin/nix build github:exo-explore/exo/$commit" &
+    "/nix/var/nix/profiles/default/bin/nix build github:exo-explore/exo/$commit" &
 done
 wait
 for host; do
   ssh -T -o BatchMode=yes -o ServerAliveInterval=30 "$host@$host" \
-    "EXO_LIBP2P_NAMESPACE=$commit /nix/var/nix/profiles/default/bin/nix run github:exo-explore/exo/$commit" &>/dev/null &
+    "/nix/var/nix/profiles/default/bin/nix run github:exo-explore/exo/$commit -- --namespace $commit" &>/dev/null &
 done
 
 for host; do

tests/run_exo_on.sh

@@ -35,7 +35,7 @@ i=0
 for host; do
   colour=${colours[i++ % 4]}
   ssh -T -o BatchMode=yes -o ServerAliveInterval=30 "$host@$host" \
-    "EXO_LIBP2P_NAMESPACE=$commit /nix/var/nix/profiles/default/bin/nix run github:exo-explore/exo/$commit" |&
+    "/nix/var/nix/profiles/default/bin/nix run github:exo-explore/exo/$commit -- --namespace $commit" |&
     awk -v p="${colour}[${host}]${reset}" '{ print p $0; fflush() }' &
 done
 

uv.lock

@@ -193,20 +193,14 @@ sdist = { url = "https://files.pythonhosted.org/packages/eb/56/b1ba7935a17738ae8
 wheels = [
     { url = "https://files.pythonhosted.org/packages/b0/1e/d22cc63332bd59b06481ceaac49d6c507598642e2230f201649058a7e704/cffi-2.0.0-cp313-cp313-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:07b271772c100085dd28b74fa0cd81c8fb1a3ba18b21e03d7c27f3436a10606b", size = 212446, upload-time = "2025-09-08T23:23:03.472Z" },
     { url = "https://files.pythonhosted.org/packages/a9/f5/a2c23eb03b61a0b8747f211eb716446c826ad66818ddc7810cc2cc19b3f2/cffi-2.0.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d48a880098c96020b02d5a1f7d9251308510ce8858940e6fa99ece33f610838b", size = 220101, upload-time = "2025-09-08T23:23:04.792Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/7f/e6647792fc5850d634695bc0e6ab4111ae88e89981d35ac269956605feba/cffi-2.0.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:f93fd8e5c8c0a4aa1f424d6173f14a892044054871c771f8566e4008eaa359d2", size = 207948, upload-time = "2025-09-08T23:23:06.127Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/1e/a5a1bd6f1fb30f22573f76533de12a00bf274abcdc55c8edab639078abb6/cffi-2.0.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:dd4f05f54a52fb558f1ba9f528228066954fee3ebe629fc1660d874d040ae5a3", size = 206422, upload-time = "2025-09-08T23:23:07.753Z" },
     { url = "https://files.pythonhosted.org/packages/98/df/0a1755e750013a2081e863e7cd37e0cdd02664372c754e5560099eb7aa44/cffi-2.0.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:c8d3b5532fc71b7a77c09192b4a5a200ea992702734a2e9279a37f2478236f26", size = 219499, upload-time = "2025-09-08T23:23:09.648Z" },
     { url = "https://files.pythonhosted.org/packages/50/e1/a969e687fcf9ea58e6e2a928ad5e2dd88cc12f6f0ab477e9971f2309b57c/cffi-2.0.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:d9b29c1f0ae438d5ee9acb31cadee00a58c46cc9c0b2f9038c6b0b3470877a8c", size = 222928, upload-time = "2025-09-08T23:23:10.928Z" },
     { url = "https://files.pythonhosted.org/packages/36/54/0362578dd2c9e557a28ac77698ed67323ed5b9775ca9d3fe73fe191bb5d8/cffi-2.0.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:6d50360be4546678fc1b79ffe7a66265e28667840010348dd69a314145807a1b", size = 221302, upload-time = "2025-09-08T23:23:12.42Z" },
     { url = "https://files.pythonhosted.org/packages/d6/43/0e822876f87ea8a4ef95442c3d766a06a51fc5298823f884ef87aaad168c/cffi-2.0.0-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:24b6f81f1983e6df8db3adc38562c83f7d4a0c36162885ec7f7b77c7dcbec97b", size = 220049, upload-time = "2025-09-08T23:23:20.853Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/89/76799151d9c2d2d1ead63c2429da9ea9d7aac304603de0c6e8764e6e8e70/cffi-2.0.0-cp314-cp314-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:12873ca6cb9b0f0d3a0da705d6086fe911591737a59f28b7936bdfed27c0d47c", size = 207793, upload-time = "2025-09-08T23:23:22.08Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/dd/3465b14bb9e24ee24cb88c9e3730f6de63111fffe513492bf8c808a3547e/cffi-2.0.0-cp314-cp314-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:d9b97165e8aed9272a6bb17c01e3cc5871a594a446ebedc996e2397a1c1ea8ef", size = 206300, upload-time = "2025-09-08T23:23:23.314Z" },
     { url = "https://files.pythonhosted.org/packages/47/d9/d83e293854571c877a92da46fdec39158f8d7e68da75bf73581225d28e90/cffi-2.0.0-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:afb8db5439b81cf9c9d0c80404b60c3cc9c3add93e114dcae767f1477cb53775", size = 219244, upload-time = "2025-09-08T23:23:24.541Z" },
     { url = "https://files.pythonhosted.org/packages/2b/0f/1f177e3683aead2bb00f7679a16451d302c436b5cbf2505f0ea8146ef59e/cffi-2.0.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:737fe7d37e1a1bffe70bd5754ea763a62a066dc5913ca57e957824b72a85e205", size = 222828, upload-time = "2025-09-08T23:23:26.143Z" },
     { url = "https://files.pythonhosted.org/packages/c6/0f/cafacebd4b040e3119dcb32fed8bdef8dfe94da653155f9d0b9dc660166e/cffi-2.0.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:38100abb9d1b1435bc4cc340bb4489635dc2f0da7456590877030c9b3d40b0c1", size = 220926, upload-time = "2025-09-08T23:23:27.873Z" },
     { url = "https://files.pythonhosted.org/packages/be/b4/c56878d0d1755cf9caa54ba71e5d049479c52f9e4afc230f06822162ab2f/cffi-2.0.0-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:7cc09976e8b56f8cebd752f7113ad07752461f48a58cbba644139015ac24954c", size = 221593, upload-time = "2025-09-08T23:23:31.91Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/0d/eb704606dfe8033e7128df5e90fee946bbcb64a04fcdaa97321309004000/cffi-2.0.0-cp314-cp314t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:92b68146a71df78564e4ef48af17551a5ddd142e5190cdf2c5624d0c3ff5b2e8", size = 209354, upload-time = "2025-09-08T23:23:33.214Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/19/3c435d727b368ca475fb8742ab97c9cb13a0de600ce86f62eab7fa3eea60/cffi-2.0.0-cp314-cp314t-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:b1e74d11748e7e98e2f426ab176d4ed720a64412b6a15054378afdb71e0f37dc", size = 208480, upload-time = "2025-09-08T23:23:34.495Z" },
     { url = "https://files.pythonhosted.org/packages/d0/44/681604464ed9541673e486521497406fadcc15b5217c3e326b061696899a/cffi-2.0.0-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:28a3a209b96630bca57cce802da70c266eb08c6e97e5afd61a75611ee6c64592", size = 221584, upload-time = "2025-09-08T23:23:36.096Z" },
     { url = "https://files.pythonhosted.org/packages/25/8e/342a504ff018a2825d395d44d63a767dd8ebc927ebda557fecdaca3ac33a/cffi-2.0.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:7553fb2090d71822f02c629afe6042c299edf91ba1bf94951165613553984512", size = 224443, upload-time = "2025-09-08T23:23:37.328Z" },
     { url = "https://files.pythonhosted.org/packages/e1/5e/b666bacbbc60fbf415ba9988324a132c9a7a0448a9a8f125074671c0f2c3/cffi-2.0.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:6c6c373cfc5c83a975506110d17457138c8c63016b563cc9ed6e056a82f13ce4", size = 223437, upload-time = "2025-09-08T23:23:38.945Z" },
@@ -312,10 +306,8 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/5c/49/498c86566a1d80e978b42f0d702795f69887005548c041636df6ae1ca64c/cryptography-46.0.3-cp311-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:01ca9ff2885f3acc98c29f1860552e37f6d7c7d013d7334ff2a9de43a449315d", size = 4450807, upload-time = "2025-10-15T23:16:56.414Z" },
     { url = "https://files.pythonhosted.org/packages/4b/0a/863a3604112174c8624a2ac3c038662d9e59970c7f926acdcfaed8d61142/cryptography-46.0.3-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:6eae65d4c3d33da080cff9c4ab1f711b15c1d9760809dad6ea763f3812d254cb", size = 4299615, upload-time = "2025-10-15T23:16:58.442Z" },
     { url = "https://files.pythonhosted.org/packages/64/02/b73a533f6b64a69f3cd3872acb6ebc12aef924d8d103133bb3ea750dc703/cryptography-46.0.3-cp311-abi3-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:e5bf0ed4490068a2e72ac03d786693adeb909981cc596425d09032d372bcc849", size = 4016800, upload-time = "2025-10-15T23:17:00.378Z" },
-    { url = "https://files.pythonhosted.org/packages/25/d5/16e41afbfa450cde85a3b7ec599bebefaef16b5c6ba4ec49a3532336ed72/cryptography-46.0.3-cp311-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:5ecfccd2329e37e9b7112a888e76d9feca2347f12f37918facbb893d7bb88ee8", size = 4984707, upload-time = "2025-10-15T23:17:01.98Z" },
     { url = "https://files.pythonhosted.org/packages/c9/56/e7e69b427c3878352c2fb9b450bd0e19ed552753491d39d7d0a2f5226d41/cryptography-46.0.3-cp311-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:a2c0cd47381a3229c403062f764160d57d4d175e022c1df84e168c6251a22eec", size = 4482541, upload-time = "2025-10-15T23:17:04.078Z" },
     { url = "https://files.pythonhosted.org/packages/78/f6/50736d40d97e8483172f1bb6e698895b92a223dba513b0ca6f06b2365339/cryptography-46.0.3-cp311-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:549e234ff32571b1f4076ac269fcce7a808d3bf98b76c8dd560e42dbc66d7d91", size = 4299464, upload-time = "2025-10-15T23:17:05.483Z" },
-    { url = "https://files.pythonhosted.org/packages/00/de/d8e26b1a855f19d9994a19c702fa2e93b0456beccbcfe437eda00e0701f2/cryptography-46.0.3-cp311-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:c0a7bb1a68a5d3471880e264621346c48665b3bf1c3759d682fc0864c540bd9e", size = 4950838, upload-time = "2025-10-15T23:17:07.425Z" },
     { url = "https://files.pythonhosted.org/packages/8f/29/798fc4ec461a1c9e9f735f2fc58741b0daae30688f41b2497dcbc9ed1355/cryptography-46.0.3-cp311-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:10b01676fc208c3e6feeb25a8b83d81767e8059e1fe86e1dc62d10a3018fa926", size = 4481596, upload-time = "2025-10-15T23:17:09.343Z" },
     { url = "https://files.pythonhosted.org/packages/15/8d/03cd48b20a573adfff7652b76271078e3045b9f49387920e7f1f631d125e/cryptography-46.0.3-cp311-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:0abf1ffd6e57c67e92af68330d05760b7b7efb243aab8377e583284dbab72c71", size = 4426782, upload-time = "2025-10-15T23:17:11.22Z" },
     { url = "https://files.pythonhosted.org/packages/fa/b1/ebacbfe53317d55cf33165bda24c86523497a6881f339f9aae5c2e13e57b/cryptography-46.0.3-cp311-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:a04bee9ab6a4da801eb9b51f1b708a1b5b5c9eb48c03f74198464c66f0d344ac", size = 4698381, upload-time = "2025-10-15T23:17:12.829Z" },
@@ -323,10 +315,8 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/c5/fd/bc1daf8230eaa075184cbbf5f8cd00ba9db4fd32d63fb83da4671b72ed8a/cryptography-46.0.3-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:39b6755623145ad5eff1dab323f4eae2a32a77a7abef2c5089a04a3d04366715", size = 4435078, upload-time = "2025-10-15T23:17:23.042Z" },
     { url = "https://files.pythonhosted.org/packages/82/98/d3bd5407ce4c60017f8ff9e63ffee4200ab3e23fe05b765cab805a7db008/cryptography-46.0.3-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:db391fa7c66df6762ee3f00c95a89e6d428f4d60e7abc8328f4fe155b5ac6e54", size = 4293460, upload-time = "2025-10-15T23:17:24.885Z" },
     { url = "https://files.pythonhosted.org/packages/26/e9/e23e7900983c2b8af7a08098db406cf989d7f09caea7897e347598d4cd5b/cryptography-46.0.3-cp314-cp314t-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:78a97cf6a8839a48c49271cdcbd5cf37ca2c1d6b7fdd86cc864f302b5e9bf459", size = 3995237, upload-time = "2025-10-15T23:17:26.449Z" },
-    { url = "https://files.pythonhosted.org/packages/91/15/af68c509d4a138cfe299d0d7ddb14afba15233223ebd933b4bbdbc7155d3/cryptography-46.0.3-cp314-cp314t-manylinux_2_28_ppc64le.whl", hash = "sha256:dfb781ff7eaa91a6f7fd41776ec37c5853c795d3b358d4896fdbb5df168af422", size = 4967344, upload-time = "2025-10-15T23:17:28.06Z" },
     { url = "https://files.pythonhosted.org/packages/ca/e3/8643d077c53868b681af077edf6b3cb58288b5423610f21c62aadcbe99f4/cryptography-46.0.3-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:6f61efb26e76c45c4a227835ddeae96d83624fb0d29eb5df5b96e14ed1a0afb7", size = 4466564, upload-time = "2025-10-15T23:17:29.665Z" },
     { url = "https://files.pythonhosted.org/packages/0e/43/c1e8726fa59c236ff477ff2b5dc071e54b21e5a1e51aa2cee1676f1c986f/cryptography-46.0.3-cp314-cp314t-manylinux_2_34_aarch64.whl", hash = "sha256:23b1a8f26e43f47ceb6d6a43115f33a5a37d57df4ea0ca295b780ae8546e8044", size = 4292415, upload-time = "2025-10-15T23:17:31.686Z" },
-    { url = "https://files.pythonhosted.org/packages/42/f9/2f8fefdb1aee8a8e3256a0568cffc4e6d517b256a2fe97a029b3f1b9fe7e/cryptography-46.0.3-cp314-cp314t-manylinux_2_34_ppc64le.whl", hash = "sha256:b419ae593c86b87014b9be7396b385491ad7f320bde96826d0dd174459e54665", size = 4931457, upload-time = "2025-10-15T23:17:33.478Z" },
     { url = "https://files.pythonhosted.org/packages/79/30/9b54127a9a778ccd6d27c3da7563e9f2d341826075ceab89ae3b41bf5be2/cryptography-46.0.3-cp314-cp314t-manylinux_2_34_x86_64.whl", hash = "sha256:50fc3343ac490c6b08c0cf0d704e881d0d660be923fd3076db3e932007e726e3", size = 4466074, upload-time = "2025-10-15T23:17:35.158Z" },
     { url = "https://files.pythonhosted.org/packages/ac/68/b4f4a10928e26c941b1b6a179143af9f4d27d88fe84a6a3c53592d2e76bf/cryptography-46.0.3-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:22d7e97932f511d6b0b04f2bfd818d73dcd5928db509460aaf48384778eb6d20", size = 4420569, upload-time = "2025-10-15T23:17:37.188Z" },
     { url = "https://files.pythonhosted.org/packages/a3/49/3746dab4c0d1979888f125226357d3262a6dd40e114ac29e3d2abdf1ec55/cryptography-46.0.3-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:d55f3dffadd674514ad19451161118fd010988540cee43d8bc20675e775925de", size = 4681941, upload-time = "2025-10-15T23:17:39.236Z" },
@@ -334,10 +324,8 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/26/42/fa8389d4478368743e24e61eea78846a0006caffaf72ea24a15159215a14/cryptography-46.0.3-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:15ab9b093e8f09daab0f2159bb7e47532596075139dd74365da52ecc9cb46c5d", size = 4440029, upload-time = "2025-10-15T23:17:49.837Z" },
     { url = "https://files.pythonhosted.org/packages/5f/eb/f483db0ec5ac040824f269e93dd2bd8a21ecd1027e77ad7bdf6914f2fd80/cryptography-46.0.3-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:46acf53b40ea38f9c6c229599a4a13f0d46a6c3fa9ef19fc1a124d62e338dfa0", size = 4297222, upload-time = "2025-10-15T23:17:51.357Z" },
     { url = "https://files.pythonhosted.org/packages/fd/cf/da9502c4e1912cb1da3807ea3618a6829bee8207456fbbeebc361ec38ba3/cryptography-46.0.3-cp38-abi3-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:10ca84c4668d066a9878890047f03546f3ae0a6b8b39b697457b7757aaf18dbc", size = 4012280, upload-time = "2025-10-15T23:17:52.964Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/8f/9adb86b93330e0df8b3dcf03eae67c33ba89958fc2e03862ef1ac2b42465/cryptography-46.0.3-cp38-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:36e627112085bb3b81b19fed209c05ce2a52ee8b15d161b7c643a7d5a88491f3", size = 4978958, upload-time = "2025-10-15T23:17:54.965Z" },
     { url = "https://files.pythonhosted.org/packages/d1/a0/5fa77988289c34bdb9f913f5606ecc9ada1adb5ae870bd0d1054a7021cc4/cryptography-46.0.3-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:1000713389b75c449a6e979ffc7dcc8ac90b437048766cef052d4d30b8220971", size = 4473714, upload-time = "2025-10-15T23:17:56.754Z" },
     { url = "https://files.pythonhosted.org/packages/14/e5/fc82d72a58d41c393697aa18c9abe5ae1214ff6f2a5c18ac470f92777895/cryptography-46.0.3-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:b02cf04496f6576afffef5ddd04a0cb7d49cf6be16a9059d793a30b035f6b6ac", size = 4296970, upload-time = "2025-10-15T23:17:58.588Z" },
-    { url = "https://files.pythonhosted.org/packages/78/06/5663ed35438d0b09056973994f1aec467492b33bd31da36e468b01ec1097/cryptography-46.0.3-cp38-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:71e842ec9bc7abf543b47cf86b9a743baa95f4677d22baa4c7d5c69e49e9bc04", size = 4940236, upload-time = "2025-10-15T23:18:00.897Z" },
     { url = "https://files.pythonhosted.org/packages/fc/59/873633f3f2dcd8a053b8dd1d38f783043b5fce589c0f6988bf55ef57e43e/cryptography-46.0.3-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:402b58fc32614f00980b66d6e56a5b4118e6cb362ae8f3fda141ba4689bd4506", size = 4472642, upload-time = "2025-10-15T23:18:02.749Z" },
     { url = "https://files.pythonhosted.org/packages/3d/39/8e71f3930e40f6877737d6f69248cf74d4e34b886a3967d32f919cc50d3b/cryptography-46.0.3-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:ef639cb3372f69ec44915fafcd6698b6cc78fbe0c2ea41be867f6ed612811963", size = 4423126, upload-time = "2025-10-15T23:18:04.85Z" },
     { url = "https://files.pythonhosted.org/packages/cd/c7/f65027c2810e14c3e7268353b1681932b87e5a48e65505d8cc17c99e36ae/cryptography-46.0.3-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:3b51b8ca4f1c6453d8829e1eb7299499ca7f313900dd4d89a24b8b87c0a780d4", size = 4686573, upload-time = "2025-10-15T23:18:06.908Z" },
@@ -377,8 +365,8 @@ dependencies = [
     { name = "hypercorn", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "loguru", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "mflux", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.6", source = { registry = "https://pypi.org/simple" }, extra = ["cpu"], marker = "sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.dev20260218+14841977", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#1484197707f35186ad3bd614357c7c47fdf86ebc" }, marker = "sys_platform == 'darwin'" },
+    { name = "mlx", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+    { name = "mlx", extra = ["cpu"], marker = "sys_platform == 'linux'" },
     { name = "mlx-lm", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "msgspec", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "openai-harmony", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -416,9 +404,9 @@ requires-dist = [
     { name = "hypercorn", specifier = ">=0.18.0" },
     { name = "loguru", specifier = ">=0.7.3" },
     { name = "mflux", specifier = "==0.15.5" },
-    { name = "mlx", marker = "sys_platform == 'darwin'", git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks" },
+    { name = "mlx", marker = "sys_platform == 'darwin'", specifier = "==0.30.6" },
     { name = "mlx", extras = ["cpu"], marker = "sys_platform == 'linux'", specifier = "==0.30.6" },
-    { name = "mlx-lm", specifier = "==0.30.7" },
+    { name = "mlx-lm", specifier = "==0.30.6" },
     { name = "msgspec", specifier = ">=0.19.0" },
     { name = "openai-harmony", specifier = ">=0.0.8" },
     { name = "pillow", specifier = ">=11.0,<12.0" },
@@ -1020,8 +1008,8 @@ dependencies = [
     { name = "fonttools", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "huggingface-hub", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "matplotlib", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.6", source = { registry = "https://pypi.org/simple" }, extra = ["cuda13"], marker = "sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.dev20260218+14841977", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#1484197707f35186ad3bd614357c7c47fdf86ebc" }, marker = "sys_platform == 'darwin'" },
+    { name = "mlx", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+    { name = "mlx", extra = ["cuda13"], marker = "sys_platform == 'linux'" },
     { name = "numpy", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "opencv-python", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "piexif", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -1048,12 +1036,18 @@ wheels = [
 name = "mlx"
 version = "0.30.6"
 source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "sys_platform == 'linux'",
+dependencies = [
+    { name = "mlx-metal", marker = "sys_platform == 'darwin'" },
 ]
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/ae/5b/e460e144a34d5529e010056cccf50b538d56ed001473bc6b246018fd58cb/mlx-0.30.6-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:ed86f8bffc174c2f259ca589ea25464c96cf69d1bb457074a2bf2ef53737e54f", size = 573515, upload-time = "2026-02-06T03:45:23.405Z" },
+    { url = "https://files.pythonhosted.org/packages/60/25/69833fefb9a3fef30b56792b1bcd022496c4fea83e45411d289b77ef7546/mlx-0.30.6-cp313-cp313-macosx_15_0_arm64.whl", hash = "sha256:c52294958269e20f300639a17c1900ca8fc737d859ddda737f9811e94bd040e5", size = 573516, upload-time = "2026-02-06T03:45:24.618Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/6a/7e7fbeebc5cb51b6a5eba96b263a6298707bcbdc059f4b0b73e088bc3dea/mlx-0.30.6-cp313-cp313-macosx_26_0_arm64.whl", hash = "sha256:b5b6636f7c49a4d86d8ec82643b972f45a144a7a9f3a967b27b2e6e22cf71e6a", size = 573592, upload-time = "2026-02-06T03:45:25.928Z" },
     { url = "https://files.pythonhosted.org/packages/93/06/280f6f2ba80520a7109730425eda0d966658793aa0d02d8be8d351f75253/mlx-0.30.6-cp313-cp313-manylinux_2_35_aarch64.whl", hash = "sha256:67e6c9e30a9faeacc209917ef5523177cf9b086914b6b5d83ff886e4294b727d", size = 622011, upload-time = "2026-02-06T03:45:28.165Z" },
     { url = "https://files.pythonhosted.org/packages/fe/35/f872afbee9c079cc69924d9e9c46f5663adb7da58cba3511db082dd307c1/mlx-0.30.6-cp313-cp313-manylinux_2_35_x86_64.whl", hash = "sha256:47db8b16fcb6f6c5a47c0bdb24ed377b41237017ac93aa6cb6aa206c9bdf82e4", size = 663650, upload-time = "2026-02-06T03:45:30.315Z" },
+    { url = "https://files.pythonhosted.org/packages/60/23/361dc7a5797634e4d7e9bdd6564c6b28f9b1246672632def2f91bf066b18/mlx-0.30.6-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:78804a89dcff4a838f7c2da72392fe87a523e95122a3c840e53df019122aad45", size = 575028, upload-time = "2026-02-06T03:45:31.549Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/69/1854484d414171586814dfbe8def95f75c4ea2c7341ba13ba8ee675f7c62/mlx-0.30.6-cp314-cp314-macosx_15_0_arm64.whl", hash = "sha256:ec13584ab069665cc7ad34a05494d9291cd623aef6ae96be48875fc87cfc25d6", size = 575026, upload-time = "2026-02-06T03:45:33.072Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/b8/3adbc441924209a7e4c568308b2a0b54bd09aee6a68db5bae85304791e54/mlx-0.30.6-cp314-cp314-macosx_26_0_arm64.whl", hash = "sha256:b2c5e8a090a753ef99a1380a4d059c983083f36198864f6df9faaf1223d083df", size = 575041, upload-time = "2026-02-06T03:45:34.814Z" },
     { url = "https://files.pythonhosted.org/packages/3f/54/9d9e06804fb2088202a2cdf60458e00b221f71420bea285720b60f9e82b5/mlx-0.30.6-cp314-cp314-manylinux_2_35_aarch64.whl", hash = "sha256:9ceddede4af0de31d1f6b3099f70e5469d60cd7c546975dedbdbeab3519cab3f", size = 624002, upload-time = "2026-02-06T03:45:36Z" },
     { url = "https://files.pythonhosted.org/packages/42/92/3140a15a50cb1f9267a6552171e1dfa577861de53e093124bc43707f2a0e/mlx-0.30.6-cp314-cp314-manylinux_2_35_x86_64.whl", hash = "sha256:4a6ffd2d16728cf95f63a1b555d7c2eaeea686a0e6b73228bd265411cb5d77a4", size = 663569, upload-time = "2026-02-06T03:45:37.242Z" },
 ]
@@ -1066,14 +1060,6 @@ cuda13 = [
     { name = "mlx-cuda-13", marker = "sys_platform == 'linux'" },
 ]
 
-[[package]]
-name = "mlx"
-version = "0.30.7.dev20260218+14841977"
-source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#1484197707f35186ad3bd614357c7c47fdf86ebc" }
-resolution-markers = [
-    "sys_platform == 'darwin'",
-]
-
 [[package]]
 name = "mlx-cpu"
 version = "0.30.6"
@@ -1100,20 +1086,30 @@ wheels = [
 
 [[package]]
 name = "mlx-lm"
-version = "0.30.7"
+version = "0.30.6"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jinja2", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.dev20260218+14841977", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#1484197707f35186ad3bd614357c7c47fdf86ebc" }, marker = "sys_platform == 'darwin'" },
+    { name = "mlx", marker = "sys_platform == 'darwin'" },
     { name = "numpy", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "protobuf", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "pyyaml", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "sentencepiece", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "transformers", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/66/0d/56542e2ae13ec6f542d3977d7cff89a205d4f6c5122e0ce23f33265f61c9/mlx_lm-0.30.7.tar.gz", hash = "sha256:e5f31ac58d9f2381f28e1ba639ff903e64f7cff1bdc245c0bc97f72264be329c", size = 275764, upload-time = "2026-02-12T18:41:11.86Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/76/cb/815deddc8699b1f694d7e1f9cbed52934c03a8b49432c8add72932bb2f0b/mlx_lm-0.30.6.tar.gz", hash = "sha256:807e042d7040268f1b19190b7eaefd8b2efbff5590a65460974ad4225b91dda1", size = 271733, upload-time = "2026-02-04T21:27:45.741Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/1e/17/a41c798a3d9cbdc47f39c6db5bba4c2cd199203ead26bf911cb03b644070/mlx_lm-0.30.7-py3-none-any.whl", hash = "sha256:17442a4bf01c4c2d3bca1e647712fe44f19890c3f1eadc8589d389e57b44b9bf", size = 386591, upload-time = "2026-02-12T18:41:10.236Z" },
+    { url = "https://files.pythonhosted.org/packages/20/5f/01d281f1fa8a1521d5936659beb4f5ab1f32b463d059263cf9d4cef969d9/mlx_lm-0.30.6-py3-none-any.whl", hash = "sha256:a7405bd581eacc4bf8209d7a6b7f23629585a0d7c6740c2a97e51fee35b3b0e1", size = 379451, upload-time = "2026-02-04T21:27:43.222Z" },
+]
+
+[[package]]
+name = "mlx-metal"
+version = "0.30.6"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f3/85/44406b521f920248fad621334d4dc15e77660a494edf890e7cbee33bf38d/mlx_metal-0.30.6-py3-none-macosx_14_0_arm64.whl", hash = "sha256:ea6d0c973def9a5b4f652cc77036237db3f88c9d0af63701d76b5fddde99b820", size = 38437818, upload-time = "2026-02-06T03:44:56.19Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/cb/10a516995f7d0c154b0d7e633c54b51e96977a86a355105b6474cfcbe0d0/mlx_metal-0.30.6-py3-none-macosx_15_0_arm64.whl", hash = "sha256:0f8cb94634d07e06a372d6ad9a090f38a18bab1ff19a140aede60eacf707bb94", size = 38433701, upload-time = "2026-02-06T03:44:59.678Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/7d/70cb272f7373c334709f210ed8420511fc9d64d05a7a646c0b3b94c29c04/mlx_metal-0.30.6-py3-none-macosx_26_0_arm64.whl", hash = "sha256:d761ae26304f2c4b454eeea7f612a56919d9e5e57dbb1dc0788f8e34aa6f41c2", size = 47718448, upload-time = "2026-02-06T03:45:03.133Z" },
 ]
 
 [[package]]