Compare commits

...

10 Commits

Author SHA1 Message Date
Evan
7b4cb00787 workin on it 2026-02-18 17:12:26 +00:00
Evan
1397ee38a5 simplify ident module 2026-02-18 17:12:20 +00:00
Evan
a808b93b7c remove nightly 2026-02-18 17:12:16 +00:00
Alex Cheema
6c322ebb72 feat: only show thinking toggle for models that support it (#1497)
## Summary
- Adds `thinking_toggle` capability to 26 model cards that support
toggling thinking mode on/off
- GPT-OSS models (20b, 120b) excluded — they always think and don't
support toggling
- Dashboard UI updated to check for `thinking_toggle` capability before
showing the toggle button

## Test plan
- [x] `uv run basedpyright` — 0 errors
- [x] `uv run ruff check` — all checks passed
- [x] `nix fmt` — 0 files changed
- [x] `uv run pytest` — 188 passed, 0 failed
- [x] Security review passed (no secrets, eval/exec, innerHTML, or dep
changes)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-18 17:05:00 +00:00
vskiwi
2ebe6216b4 feat: add explicit --offline mode for air-gapped clusters (#1525)
## Motivation

Closes #1510

There is currently no reliable way to run exo on an air-gapped or offline cluster where models are pre-staged on local disks. The two existing mechanisms — `--no-downloads` and `HF_HUB_OFFLINE=1` — each cover only a subset of the problem:

1. **`--no-downloads` blocks model loading**: When passed, `DownloadCoordinator` is not created. No `NodeDownloadProgress` events are ever emitted, so `_model_needs_download()` in `plan.py` perpetually returns `DownloadModel`, short-circuiting `_load_model()` and preventing the model from ever being loaded.

2. **`HF_HUB_OFFLINE=1` doesn't cover exo's aiohttp code**: exo's download pipeline primarily uses raw `aiohttp` for HTTP operations (file list fetching, file downloads, HEAD verification), not the `huggingface_hub` library. These calls will attempt connections and time out on air-gapped networks.

3. **`skip_internet` is not propagated to `download_file_with_retry()`**: Even when `internet_connection = False`, the `_download_file()` function still makes HTTP HEAD calls via `file_meta()` to verify local files and unconditionally attempts downloads for missing files.

## Changes

### `src/exo/main.py`
- Add `--offline` flag to `Args` with env var detection (`EXO_OFFLINE=1`, `HF_HUB_OFFLINE=1`)
- Pass `offline` to `DownloadCoordinator` at creation and re-creation (election loop)

### `src/exo/download/coordinator.py`
- Add `offline: bool = False` field
- In offline mode: set `internet_connection = False` immediately in `__post_init__`, skip `_test_internet_connection()` ping (avoids 3s timeout), skip `_check_internet_connection` periodic loop
- In `_start_download()`: if model is not fully available locally, emit `DownloadFailed` with clear message instead of starting a download task

### `src/exo/download/download_utils.py`
- Add `skip_internet: bool` parameter to `download_file_with_retry()` and `_download_file()`
- When `skip_internet=True` in `_download_file()`: return local file immediately without HTTP HEAD verification; raise `FileNotFoundError` for missing files
- Propagate `skip_internet` from `download_shard()` to `download_file_with_retry()`

### `src/exo/download/tests/test_offline_mode.py` (new)
- 8 tests covering `_download_file`, `download_file_with_retry`, and `fetch_file_list_with_cache` in offline mode

## Why It Works

Unlike `--no-downloads` which disables `DownloadCoordinator` entirely, `--offline` keeps the coordinator running in a restricted mode. The existing `_emit_existing_download_progress()` disk scanner still runs every 60 seconds, emitting `DownloadCompleted` events for pre-staged models. These events flow through the event-sourcing pipeline and populate `state.downloads`, which unblocks `_model_needs_download()` in `plan.py` — no changes to the planning logic required.

```
--offline flag
  → DownloadCoordinator (offline mode)
    → Skip 1.1.1.1 ping, internet_connection = False
    → _emit_existing_download_progress scans disk
      → Emits DownloadCompleted for pre-staged models
        → _model_needs_download sees DownloadCompleted
          → _load_model proceeds normally
```

## Test Plan

### Automated Testing
- `ruff check` — passes
- 8 new tests in `test_offline_mode.py` — all pass
- 11 existing download tests in `test_download_verification.py` — all pass (no regressions)

### Manual Testing
1. Pre-stage a model on disk (e.g., `~/.exo/models/mlx-community--Qwen3-0.6B-4bit/`)
2. Start exo with `--offline` (or `EXO_OFFLINE=1`)
3. Place an instance via API or dashboard
4. Verify: model loads into memory and inference works without any network calls

### Environment
- macOS (Apple Silicon), multi-node cluster with Thunderbolt interconnect
- Models pre-staged via rsync / NFS mount
2026-02-18 16:18:09 +00:00
ciaranbor
f54c80b121 Ciaran/image edit api (#1500)
## Motivation

- Image editing previously ignored input image dimensions, always
defaulting to 1024x1024
- Size dropdown was hidden in edit mode, giving users no control over
output dimensions
- Portrait/landscape presets used non-standard aspect ratios (1024x1365
/ 1365x1024)

## Changes

- Added "auto" size option that uses input image dimensions for edits,
defaults to 1024x1024 for generation
- Introduced ImageSize Literal type and normalize_image_size() validator
(replaces raw str size fields)
  - Updated portrait/landscape presets to standard 1024x1536 / 1536x1024
  - Made size selector visible in edit mode (previously hidden)
  - Default size changed from "1024x1024" to "auto"

## Why It Works

- "auto" reads actual input image dimensions via PIL at generation time,
so edits preserve the original aspect ratio
- Pydantic field_validator on both ImageGenerationTaskParams and
ImageEditsTaskParams normalizes None → "auto", keeping the API
backward-compatible

## Test Plan

### Manual Testing

- Verify image edits output at the input image's native resolution when
size is "auto"
- Verify size dropdown appears and works in both generate and edit modes
2026-02-18 16:05:39 +00:00
rltakashige
48b8f86395 Add support for GLM 5 (#1526)
## Motivation

Add GLM 5 support in favor of #1513 

## Changes

<!-- Describe what you changed in detail -->

## Why It Works

<!-- Explain why your approach solves the problem -->

## Test Plan

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

### Automated Testing
<!-- Describe changes to automated tests, or how existing tests cover
this change -->
<!-- - -->
2026-02-18 14:04:06 +00:00
Evan
5cbd6377a2 prioritize official model cards over custom model cards
our old model card search path would override official model cards with
custom model cards - our packaged model cards should always be the
default here
2026-02-18 13:20:05 +00:00
Evan Quiney
8f01523ddb remove dead code (#1496) 2026-02-18 11:43:27 +00:00
Alex Cheema
3addeadea8 Update mlx-lm to 0.30.7 (#1520)
## Summary
- Bumps `mlx-lm` from 0.30.6 to 0.30.7 in `pyproject.toml` and `uv.lock`

## Test plan
- [x] `uv lock` resolves successfully
- [x] `basedpyright` — no new errors (63 pre-existing in unrelated
`test_tool_call_tracker.py`)
- [x] `ruff check` — all checks passed
- [x] `nix fmt` — no formatting changes
- [x] `pytest` — 188 passed, 1 skipped

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-18 11:14:23 +00:00
74 changed files with 931 additions and 1633 deletions

View File

@@ -0,0 +1,46 @@
"""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: ...

137
Cargo.lock generated
View File

@@ -141,12 +141,6 @@ version = "0.3.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "76a2e8124351fda1ef8aaaa3bbd7ebbcb486bbcd4225aca0aa0d84bb2db8fecb"
[[package]]
name = "arrayvec"
version = "0.7.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7c02d123df017efcdfbd739ef81735b36c5ba83ec3c59c80a9d7ecc718f92e50"
[[package]]
name = "asn1-rs"
version = "0.7.1"
@@ -304,19 +298,6 @@ version = "1.8.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "55248b47b0caf0546f7988906588779981c43bb1bc9d0c44087278f80cdb44ba"
[[package]]
name = "bigdecimal"
version = "0.4.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "560f42649de9fa436b73517378a147ec21f6c997a546581df4b4b31677828934"
dependencies = [
"autocfg",
"libm",
"num-bigint",
"num-integer",
"num-traits",
]
[[package]]
name = "bimap"
version = "0.6.3"
@@ -516,15 +497,6 @@ version = "0.4.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2f421161cb492475f1661ddc9815a745a1c894592070661180fdec3d4872e9c3"
[[package]]
name = "convert_case"
version = "0.10.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "633458d4ef8c78b72454de2d54fd6ab2e60f9e02be22f3c6104cdc8a4e0fceb9"
dependencies = [
"unicode-segmentation",
]
[[package]]
name = "core-foundation"
version = "0.9.4"
@@ -746,29 +718,6 @@ dependencies = [
"powerfmt",
]
[[package]]
name = "derive_more"
version = "2.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "10b768e943bed7bf2cab53df09f4bc34bfd217cdb57d971e769874c9a6710618"
dependencies = [
"derive_more-impl",
]
[[package]]
name = "derive_more-impl"
version = "2.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6d286bfdaf75e988b4a78e013ecd79c581e06399ab53fbacd2d916c2f904f30b"
dependencies = [
"convert_case",
"proc-macro2",
"quote",
"rustc_version",
"syn 2.0.111",
"unicode-xid",
]
[[package]]
name = "digest"
version = "0.10.7"
@@ -939,22 +888,17 @@ name = "exo_pyo3_bindings"
version = "0.0.1"
dependencies = [
"delegate",
"derive_more",
"env_logger",
"extend",
"futures",
"impl-trait-for-tuples",
"futures-lite",
"libp2p",
"log",
"networking",
"once_cell",
"pin-project",
"pyo3",
"pyo3-async-runtimes",
"pyo3-log",
"pyo3-stub-gen",
"thiserror 2.0.17",
"thread_local",
"tokio",
"util",
]
@@ -970,6 +914,12 @@ dependencies = [
"syn 2.0.111",
]
[[package]]
name = "fastrand"
version = "2.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be"
[[package]]
name = "ff"
version = "0.13.1"
@@ -1078,7 +1028,10 @@ 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",
]
@@ -1640,17 +1593,6 @@ dependencies = [
"xmltree",
]
[[package]]
name = "impl-trait-for-tuples"
version = "0.2.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a0eb5a3343abf848c0984fe4604b2b105da9539376e24fc0a3b0007411ae4fd9"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.111",
]
[[package]]
name = "indexmap"
version = "2.12.1"
@@ -1829,12 +1771,6 @@ version = "0.2.178"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "37c93d8daa9d8a012fd8ab92f088405fb202ea0b6ab73ee2482ae66af4f42091"
[[package]]
name = "libm"
version = "0.2.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f9fbbcab51052fe104eb5e5d351cf728d30a5be1fe14d9be8a3b097481fb97de"
[[package]]
name = "libp2p"
version = "0.56.0"
@@ -2824,16 +2760,14 @@ name = "networking"
version = "0.0.1"
dependencies = [
"delegate",
"derive_more",
"either",
"extend",
"futures",
"futures-lite",
"futures-timer",
"impl-trait-for-tuples",
"keccak-const",
"libp2p",
"log",
"thiserror 2.0.17",
"pin-project",
"tokio",
"tracing-subscriber",
"util",
@@ -2918,17 +2852,6 @@ dependencies = [
"num-traits",
]
[[package]]
name = "num-rational"
version = "0.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f83d14da390562dca69fc84082e73e548e1ad308d24accdedd2720017cb37824"
dependencies = [
"num-bigint",
"num-integer",
"num-traits",
]
[[package]]
name = "num-traits"
version = "0.2.19"
@@ -3279,28 +3202,14 @@ version = "0.27.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ab53c047fcd1a1d2a8820fe84f05d6be69e9526be40cb03b73f86b6b03e6d87d"
dependencies = [
"bigdecimal",
"either",
"hashbrown 0.16.1",
"indexmap",
"indoc",
"inventory",
"libc",
"lock_api",
"memoffset",
"num-bigint",
"num-complex",
"num-rational",
"num-traits",
"once_cell",
"ordered-float",
"parking_lot",
"portable-atomic",
"pyo3-build-config",
"pyo3-ffi",
"pyo3-macros",
"rust_decimal",
"smallvec",
"unindent",
]
@@ -3741,16 +3650,6 @@ dependencies = [
"tokio",
]
[[package]]
name = "rust_decimal"
version = "1.39.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "35affe401787a9bd846712274d97654355d21b2a2c092a3139aabe31e9022282"
dependencies = [
"arrayvec",
"num-traits",
]
[[package]]
name = "rustc-hash"
version = "1.1.0"
@@ -4615,24 +4514,12 @@ version = "1.0.22"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9312f7c4f6ff9069b165498234ce8be658059c6728633667c526e27dc2cf1df5"
[[package]]
name = "unicode-segmentation"
version = "1.12.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f6ccf251212114b54433ec949fd6a7841275f9ada20dddd2f29e9ceea4501493"
[[package]]
name = "unicode-width"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b4ac048d71ede7ee76d585517add45da530660ef4390e49b098733c6e897f254"
[[package]]
name = "unicode-xid"
version = "0.2.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ebc1c04c71510c7f702b52b7c350734c9ff1295c464a03335b00bb84fc54f853"
[[package]]
name = "unicode_names2"
version = "1.3.0"

View File

@@ -26,49 +26,20 @@ opt-level = 3
networking = { path = "rust/networking" }
util = { path = "rust/util" }
# Proc-macro authoring tools
syn = "2.0"
quote = "1.0"
proc-macro2 = "1.0"
darling = "0.20"
# Macro dependecies
extend = "1.2"
delegate = "0.13"
impl-trait-for-tuples = "0.2"
clap = "4.5"
derive_more = { version = "2.0.1", features = ["display"] }
pin-project = "1"
# Utility dependencies
itertools = "0.14"
thiserror = "2"
internment = "0.8"
recursion = "0.5"
regex = "1.11"
once_cell = "1.21"
thread_local = "1.1"
bon = "3.4"
generativity = "1.1"
anyhow = "1.0"
keccak-const = "0.2"
# Functional generics/lenses frameworks
frunk_core = "0.4"
frunk = "0.4"
frunk_utils = "0.2"
frunk-enum-core = "0.3"
# Async dependencies
tokio = "1.46"
futures = "0.3"
futures-util = "0.3"
futures-lite = "2.6.1"
futures-timer = "3.0"
# Data structures
either = "1.15"
ordered-float = "5.0"
ahash = "0.8"
# Tracing/logging
log = "0.4"

View File

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

View File

@@ -59,13 +59,14 @@
}
const sizeOptions: ImageGenerationParams["size"][] = [
"auto",
"512x512",
"768x768",
"1024x1024",
"1024x768",
"768x1024",
"1024x1365",
"1365x1024",
"1024x1536",
"1536x1024",
];
const qualityOptions: ImageGenerationParams["quality"][] = [
@@ -176,92 +177,90 @@
<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 (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
<!-- 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'
: ''}"
>
<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'
: ''}"
{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"
>
{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>
<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}
</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 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}
{/if}
</div>
<!-- Quality -->
<div class="flex items-center gap-1.5">
@@ -311,7 +310,7 @@
<!-- 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 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 overflow-x-hidden min-w-max"
style="bottom: calc(100vh - {qualityDropdownPosition()
.top}px + 4px); left: {qualityDropdownPosition().left}px;"
>

View File

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

View File

@@ -74,7 +74,6 @@
perSystem =
{ config, self', inputs', pkgs, lib, system, ... }:
let
fenixToolchain = inputs'.fenix.packages.complete;
# Use pinned nixpkgs for swift-format (swift is broken on x86_64-linux in newer nixpkgs)
pkgsSwift = import inputs.nixpkgs-swift { inherit system; };
in

View File

@@ -41,7 +41,7 @@ let
mlx = stdenv.mkDerivation rec {
pname = "mlx";
version = let v = "0.30.7.dev20260217+50487b41"; in
version = let v = "0.30.7.dev20260218+14841977"; 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;
@@ -49,8 +49,8 @@ let
src = fetchFromGitHub {
owner = "rltakashige";
repo = "mlx-jaccl-fix-small-recv";
rev = "50487b4141f3c951122655db3b83df5146c1fbeb";
hash = "sha256-IL4a9vMX5nocgJU1WG4zE8hArHkHJtnh4sdYh3od5zU=";
rev = "1484197707f35186ad3bd614357c7c47fdf86ebc";
hash = "sha256-FupCMoK/SF/ldfKuvMSAKECcOP8c+ANgkQlPZttDsLk=";
};
patches = [

View File

@@ -19,7 +19,7 @@ dependencies = [
"anyio==4.11.0",
"mlx; sys_platform == 'darwin'",
"mlx[cpu]==0.30.6; sys_platform == 'linux'",
"mlx-lm==0.30.6",
"mlx-lm==0.30.7",
"tiktoken>=0.12.0", # required for kimi k2 tokenizer
"hypercorn>=0.18.0",
"openai-harmony>=0.0.8",

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -0,0 +1,12 @@
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

View File

@@ -0,0 +1,12 @@
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

View File

@@ -0,0 +1,12 @@
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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -25,17 +25,17 @@ workspace = true
networking = { workspace = true }
# interop
pyo3 = { version = "0.27.1", features = [
# "abi3-py311", # tells pyo3 (and maturin) to build using the stable ABI with minimum Python version 3.11
"nightly", # enables better-supported GIL integration
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)
"multiple-pymethods", # allows multiple #[pymethods] sections per class
# "multiple-pymethods", # allows multiple #[pymethods] sections per class
# integrations with other libraries
"arc_lock", "bigdecimal", "either", "hashbrown", "indexmap", "num-bigint", "num-complex", "num-rational",
"ordered-float", "rust_decimal", "smallvec",
# "arc_lock", "bigdecimal", "either", "hashbrown", "indexmap", "num-bigint", "num-complex", "num-rational",
# "ordered-float", "rust_decimal", "smallvec",
# "anyhow", "chrono", "chrono-local", "chrono-tz", "eyre", "jiff-02", "lock_api", "parking-lot", "time", "serde",
] }
pyo3-stub-gen = { version = "0.17.2" }
@@ -45,33 +45,18 @@ pyo3-log = "0.13.2"
# macro dependencies
extend = { workspace = true }
delegate = { workspace = true }
impl-trait-for-tuples = { workspace = true }
derive_more = { workspace = true }
pin-project = { workspace = true }
# async runtime
tokio = { workspace = true, features = ["full", "tracing"] }
futures = { workspace = true }
futures-lite = { workspace = true }
# utility dependencies
once_cell = "1.21.3"
thread_local = "1.1.9"
util = { workspace = true }
thiserror = { workspace = true }
#internment = { workspace = true }
#recursion = { workspace = true }
#generativity = { workspace = true }
#itertools = { workspace = true }
# 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"] }
pin-project = "1.1.10"

View File

@@ -19,7 +19,7 @@ class ConnectionUpdate:
Whether this is a connection or disconnection event
"""
@property
def peer_id(self) -> PeerId:
def peer_id(self) -> builtins.str:
r"""
Identity of the peer that we have connected to or disconnected from.
"""
@@ -40,92 +40,22 @@ class Keypair:
Identity keypair of a node.
"""
@staticmethod
def generate_ed25519() -> Keypair:
def generate() -> Keypair:
r"""
Generate a new Ed25519 keypair.
"""
@staticmethod
def generate_ecdsa() -> Keypair:
def from_bytes(bytes: bytes) -> Keypair:
r"""
Generate a new ECDSA keypair.
"""
@staticmethod
def generate_secp256k1() -> Keypair:
r"""
Generate a new Secp256k1 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`.
"""
@typing.final
class Multiaddr:
r"""
Representation of a Multiaddr.
"""
@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.
Construct an Ed25519 keypair from secret key bytes
"""
def to_bytes(self) -> bytes:
r"""
Return a copy of this [`Multiaddr`]'s byte representation.
Get the secret key bytes underlying the keypair
"""
def to_string(self) -> builtins.str:
def to_node_id(self) -> builtins.str:
r"""
Convert a Multiaddr to a string.
Convert the `Keypair` into the corresponding `PeerId` string, which we use as our NodeId.
"""
@typing.final
@@ -180,37 +110,6 @@ class NoPeersSubscribedToTopicError(builtins.Exception):
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"""

View File

@@ -2,11 +2,10 @@
//!
use pin_project::pin_project;
use pyo3::marker::Ungil;
use pyo3::prelude::*;
use std::{
future::Future,
pin::{Pin, pin},
pin::Pin,
task::{Context, Poll},
};
@@ -26,15 +25,13 @@ where
impl<F> Future for AllowThreads<F>
where
F: Future + Ungil,
F::Output: Ungil,
F: Future + Send,
F::Output: Send,
{
type Output = F::Output;
fn poll(self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Self::Output> {
let waker = cx.waker();
Python::with_gil(|py| {
py.allow_threads(|| self.project().0.poll(&mut Context::from_waker(waker)))
})
Python::attach(|py| py.detach(|| self.project().0.poll(&mut Context::from_waker(waker))))
}
}

View File

@@ -1,240 +0,0 @@
//! This module exists to hold examples of some pyo3 patterns that may be too complex to
//! re-create from scratch, but too inhomogenous to create an abstraction/wrapper around.
//!
//! Pattern examples include:
//! - Async task handles: with GC-integrated cleanup
//! - Sync/async callbacks from python: with propper eventloop handling
//!
//! Mutability pattern: https://pyo3.rs/v0.26.0/async-await.html#send--static-constraint
//! - Store mutable fields in tokio's `Mutex<T>`
//! - For async code: take `&self` and `.lock().await`
//! - For sync code: take `&mut self` and `.get_mut()`
use crate::ext::{PyResultExt as _, ResultExt as _, TokioRuntimeExt as _};
use futures::FutureExt as _;
use futures::future::BoxFuture;
use pyo3::exceptions::PyRuntimeError;
use pyo3::prelude::{PyModule, PyModuleMethods as _};
use pyo3::{
Bound, Py, PyAny, PyErr, PyResult, PyTraverseError, PyVisit, Python, pyclass, pymethods,
};
use std::time::Duration;
use tokio::sync::mpsc;
use tokio::sync::mpsc::error::TryRecvError;
fn needs_tokio_runtime() {
tokio::runtime::Handle::current();
}
type SyncCallback = Box<dyn Fn() + Send + Sync>;
type AsyncCallback = Box<dyn Fn() -> BoxFuture<'static, ()> + Send + Sync>;
enum AsyncTaskMessage {
SyncCallback(SyncCallback),
AsyncCallback(AsyncCallback),
}
async fn async_task(
sender: mpsc::UnboundedSender<()>,
mut receiver: mpsc::UnboundedReceiver<AsyncTaskMessage>,
) {
log::info!("RUST: async task started");
// task state
let mut interval = tokio::time::interval(Duration::from_secs(1));
let mut sync_cbs: Vec<SyncCallback> = vec![];
let mut async_cbs: Vec<AsyncCallback> = vec![];
loop {
tokio::select! {
// handle incoming messages from task-handle
message = receiver.recv() => {
// handle closed channel by exiting
let Some(message) = message else {
log::info!("RUST: channel closed");
break;
};
// dispatch incoming event
match message {
AsyncTaskMessage::SyncCallback(cb) => {
sync_cbs.push(cb);
}
AsyncTaskMessage::AsyncCallback(cb) => {
async_cbs.push(cb);
}
}
}
// handle all other events
_ = interval.tick() => {
log::info!("RUST: async task tick");
// call back all sync callbacks
for cb in &sync_cbs {
cb();
}
// call back all async callbacks
for cb in &async_cbs {
cb().await;
}
// send event on unbounded channel
sender.send(()).expect("handle receiver cannot be closed/dropped");
}
}
}
log::info!("RUST: async task stopped");
}
// #[gen_stub_pyclass]
#[pyclass(name = "AsyncTaskHandle")]
#[derive(Debug)]
struct PyAsyncTaskHandle {
sender: Option<mpsc::UnboundedSender<AsyncTaskMessage>>,
receiver: mpsc::UnboundedReceiver<()>,
}
#[allow(clippy::expect_used)]
impl PyAsyncTaskHandle {
const fn sender(&self) -> &mpsc::UnboundedSender<AsyncTaskMessage> {
self.sender
.as_ref()
.expect("The sender should only be None after de-initialization.")
}
const fn sender_mut(&mut self) -> &mpsc::UnboundedSender<AsyncTaskMessage> {
self.sender
.as_mut()
.expect("The sender should only be None after de-initialization.")
}
const fn new(
sender: mpsc::UnboundedSender<AsyncTaskMessage>,
receiver: mpsc::UnboundedReceiver<()>,
) -> Self {
Self {
sender: Some(sender),
receiver,
}
}
}
// #[gen_stub_pymethods]
#[pymethods]
impl PyAsyncTaskHandle {
#[new]
fn py_new(py: Python<'_>) -> PyResult<Self> {
use pyo3_async_runtimes::tokio::get_runtime;
// create communication channel TOWARDS our task
let (h_sender, t_receiver) = mpsc::unbounded_channel::<AsyncTaskMessage>();
// create communication channel FROM our task
let (t_sender, h_receiver) = mpsc::unbounded_channel::<()>();
// perform necessary setup within tokio context - or it crashes
let () = get_runtime().block_on(async { needs_tokio_runtime() });
// spawn tokio task with this thread's task-locals - without this, async callbacks on the new threads will not work!!
_ = get_runtime().spawn_with_scope(py, async move {
async_task(t_sender, t_receiver).await;
});
Ok(Self::new(h_sender, h_receiver))
}
/// NOTE: exceptions in callbacks are silently ignored until end of execution
fn add_sync_callback(
&self,
// #[gen_stub(override_type(
// type_repr="collections.abc.Callable[[], None]",
// imports=("collections.abc")
// ))]
callback: Py<PyAny>,
) -> PyResult<()> {
// blocking call to async method -> can do non-blocking if needed
self.sender()
.send(AsyncTaskMessage::SyncCallback(Box::new(move || {
_ = Python::with_gil(|py| callback.call0(py).write_unraisable_with(py));
})))
.pyerr()?;
Ok(())
}
/// NOTE: exceptions in callbacks are silently ignored until end of execution
fn add_async_callback(
&self,
// #[gen_stub(override_type(
// type_repr="collections.abc.Callable[[], collections.abc.Awaitable[None]]",
// imports=("collections.abc")
// ))]
callback: Py<PyAny>,
) -> PyResult<()> {
// blocking call to async method -> can do non-blocking if needed
self.sender()
.send(AsyncTaskMessage::AsyncCallback(Box::new(move || {
let c = Python::with_gil(|py| callback.clone_ref(py));
async move {
if let Some(f) = Python::with_gil(|py| {
let coroutine = c.call0(py).write_unraisable_with(py)?;
pyo3_async_runtimes::tokio::into_future(coroutine.into_bound(py))
.write_unraisable_with(py)
}) {
_ = f.await.write_unraisable();
}
}
.boxed()
})))
.pyerr()?;
Ok(())
}
async fn receive_unit(&mut self) -> PyResult<()> {
self.receiver
.recv()
.await
.ok_or(PyErr::new::<PyRuntimeError, _>(
"cannot receive unit on closed channel",
))
}
fn drain_units(&mut self) -> PyResult<i32> {
let mut cnt = 0;
loop {
match self.receiver.try_recv() {
Err(TryRecvError::Disconnected) => {
return Err(PyErr::new::<PyRuntimeError, _>(
"cannot receive unit on closed channel",
));
}
Err(TryRecvError::Empty) => return Ok(cnt),
Ok(()) => {
cnt += 1;
continue;
}
}
}
}
// #[gen_stub(skip)]
const fn __traverse__(&self, _visit: PyVisit<'_>) -> Result<(), PyTraverseError> {
Ok(()) // This is needed purely so `__clear__` can work
}
// #[gen_stub(skip)]
fn __clear__(&mut self) {
// TODO: may or may not need to await a "kill-signal" oneshot channel message,
// to ensure that the networking task is done BEFORE exiting the clear function...
// but this may require GIL?? and it may not be safe to call GIL here??
self.sender = None; // Using Option<T> as a trick to force `sender` channel to be dropped
}
}
pub fn examples_submodule(m: &Bound<'_, PyModule>) -> PyResult<()> {
m.add_class::<PyAsyncTaskHandle>()?;
Ok(())
}

View File

@@ -0,0 +1,47 @@
use crate::ext::ResultExt as _;
use libp2p::identity::Keypair;
use pyo3::types::{PyBytes, PyBytesMethods};
use pyo3::{Bound, PyResult, Python, pyclass, pymethods};
use pyo3_stub_gen::derive::{gen_stub_pyclass, gen_stub_pymethods};
/// Identity keypair of a node.
#[gen_stub_pyclass]
#[pyclass(name = "Keypair", frozen)]
#[repr(transparent)]
pub struct PyKeypair(pub Keypair);
#[gen_stub_pymethods]
#[pymethods]
#[allow(clippy::needless_pass_by_value)]
impl PyKeypair {
/// Generate a new Ed25519 keypair.
#[staticmethod]
fn generate() -> Self {
Self(Keypair::generate_ed25519())
}
/// Construct an Ed25519 keypair from secret key bytes
#[staticmethod]
fn from_bytes(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
let mut bytes = Vec::from(bytes.as_bytes());
Ok(Self(Keypair::ed25519_from_bytes(&mut bytes).pyerr()?))
}
/// Get the secret key bytes underlying the keypair
fn to_bytes<'py>(&self, py: Python<'py>) -> PyResult<Bound<'py, PyBytes>> {
let bytes = self
.0
.clone()
.try_into_ed25519()
.expect("we only use ed25519 keys")
.secret()
.as_ref()
.to_vec();
Ok(PyBytes::new(py, &bytes))
}
/// Convert the `Keypair` into the corresponding `PeerId` string, which we use as our NodeId.
fn to_node_id(&self) -> String {
self.0.public().to_peer_id().to_base58()
}
}

View File

@@ -4,28 +4,14 @@
//!
//!
// enable Rust-unstable features for convenience
#![feature(trait_alias)]
#![feature(tuple_trait)]
#![feature(unboxed_closures)]
// #![feature(stmt_expr_attributes)]
// #![feature(assert_matches)]
// #![feature(async_fn_in_dyn_trait)]
// #![feature(async_for_loop)]
// #![feature(auto_traits)]
// #![feature(negative_impls)]
extern crate core;
mod allow_threading;
mod examples;
pub(crate) mod networking;
pub(crate) mod pylibp2p;
mod ident;
mod networking;
use crate::ident::PyKeypair;
use crate::networking::networking_submodule;
use crate::pylibp2p::ident::ident_submodule;
use crate::pylibp2p::multiaddr::multiaddr_submodule;
use pyo3::prelude::PyModule;
use pyo3::prelude::*;
use pyo3::types::PyModuleMethods;
use pyo3::{Bound, PyResult, pyclass, pymodule};
use pyo3_stub_gen::define_stub_info_gatherer;
@@ -34,24 +20,11 @@ pub(crate) mod r#const {
pub const MPSC_CHANNEL_SIZE: usize = 1024;
}
/// Namespace for all the type/trait aliases used by this crate.
pub(crate) mod alias {
use std::error::Error;
use std::marker::Tuple;
pub trait SendFn<Args: Tuple + Send + 'static, Output> =
Fn<Args, Output = Output> + Send + 'static;
pub type AnyError = Box<dyn Error + Send + Sync + 'static>;
pub type AnyResult<T> = Result<T, AnyError>;
}
/// Namespace for crate-wide extension traits/methods
pub(crate) mod ext {
use crate::allow_threading::AllowThreads;
use extend::ext;
use pyo3::exceptions::{PyConnectionError, PyRuntimeError};
use pyo3::marker::Ungil;
use pyo3::types::PyBytes;
use pyo3::{Py, PyErr, PyResult, Python};
use tokio::runtime::Runtime;
@@ -62,7 +35,7 @@ pub(crate) mod ext {
#[ext(pub, name = ByteArrayExt)]
impl [u8] {
fn pybytes(&self) -> Py<PyBytes> {
Python::with_gil(|py| PyBytes::new(py, self).unbind())
Python::attach(|py| PyBytes::new(py, self).unbind())
}
}
@@ -98,7 +71,7 @@ pub(crate) mod ext {
#[ext(pub, name = PyResultExt)]
impl<T> PyResult<T> {
fn write_unraisable(self) -> Option<T> {
Python::with_gil(|py| self.write_unraisable_with(py))
Python::attach(|py| self.write_unraisable_with(py))
}
fn write_unraisable_with(self, py: Python<'_>) -> Option<T> {
@@ -175,24 +148,6 @@ pub(crate) mod ext {
}
}
pub(crate) mod private {
use std::marker::Sized;
/// Sealed traits support
pub trait Sealed {}
impl<T: ?Sized> Sealed for T {}
}
/// A wrapper around [`Py`] that implements [`Clone`] using [`Python::with_gil`].
#[repr(transparent)]
pub(crate) struct ClonePy<T>(pub Py<T>);
impl<T> Clone for ClonePy<T> {
fn clone(&self) -> Self {
Python::with_gil(|py| Self(self.0.clone_ref(py)))
}
}
/// 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.
@@ -204,8 +159,7 @@ fn main_module(m: &Bound<'_, PyModule>) -> PyResult<()> {
// TODO: for now this is all NOT a submodule, but figure out how to make the submodule system
// work with maturin, where the types generate correctly, in the right folder, without
// too many importing issues...
ident_submodule(m)?;
multiaddr_submodule(m)?;
m.add_class::<PyKeypair>()?;
networking_submodule(m)?;
// top-level constructs

View File

@@ -1,31 +1,24 @@
#![allow(
clippy::multiple_inherent_impl,
clippy::unnecessary_wraps,
clippy::unused_self,
clippy::needless_pass_by_value
)]
use crate::r#const::MPSC_CHANNEL_SIZE;
use crate::ext::{ByteArrayExt as _, FutureExt, PyErrExt as _};
use crate::ext::{ResultExt as _, TokioMpscReceiverExt as _, TokioMpscSenderExt as _};
use crate::ext::{ResultExt as _, TokioMpscSenderExt as _};
use crate::ident::PyKeypair;
use crate::networking::exception::{PyAllQueuesFullError, PyNoPeersSubscribedToTopicError};
use crate::pyclass;
use crate::pylibp2p::ident::{PyKeypair, PyPeerId};
use libp2p::futures::StreamExt as _;
use libp2p::gossipsub::{IdentTopic, Message, MessageId, PublishError};
use libp2p::swarm::SwarmEvent;
use libp2p::{gossipsub, mdns};
use networking::discovery;
use networking::swarm::create_swarm;
use futures_lite::StreamExt as _;
use libp2p::gossipsub::PublishError;
use networking::swarm::{FromSwarm, Swarm, ToSwarm, create_swarm};
use pyo3::exceptions::PyRuntimeError;
use pyo3::prelude::{PyModule, PyModuleMethods as _};
use pyo3::types::PyBytes;
use pyo3::{Bound, Py, PyErr, PyResult, PyTraverseError, PyVisit, Python, pymethods};
use pyo3_stub_gen::derive::{gen_stub_pyclass, gen_stub_pyclass_enum, gen_stub_pymethods};
use std::net::IpAddr;
use pyo3::{Bound, Py, PyErr, PyResult, Python, pymethods};
use pyo3_stub_gen::derive::{
gen_stub_pyclass, gen_stub_pyclass_complex_enum, gen_stub_pyclass_enum, gen_stub_pymethods,
};
use tokio::sync::{Mutex, mpsc, oneshot};
mod exception {
use pyo3::types::PyTuple;
use pyo3::{PyErrArguments, exceptions::PyException, prelude::*};
use pyo3::{exceptions::PyException, prelude::*};
use pyo3_stub_gen::derive::*;
#[gen_stub_pyclass]
@@ -119,7 +112,7 @@ struct PyConnectionUpdate {
/// Identity of the peer that we have connected to or disconnected from.
#[pyo3(get)]
peer_id: PyPeerId,
peer_id: String,
/// Remote connection's IPv4 address.
#[pyo3(get)]
@@ -130,206 +123,45 @@ struct PyConnectionUpdate {
remote_tcp_port: u16,
}
enum ToTask {
GossipsubSubscribe {
topic: String,
result_tx: oneshot::Sender<PyResult<bool>>,
},
GossipsubUnsubscribe {
topic: String,
result_tx: oneshot::Sender<bool>,
},
GossipsubPublish {
topic: String,
data: Vec<u8>,
result_tx: oneshot::Sender<PyResult<MessageId>>,
},
}
#[allow(clippy::enum_glob_use)]
async fn networking_task(
mut swarm: networking::swarm::Swarm,
mut to_task_rx: mpsc::Receiver<ToTask>,
connection_update_tx: mpsc::Sender<PyConnectionUpdate>,
gossipsub_message_tx: mpsc::Sender<(String, Vec<u8>)>,
) {
use SwarmEvent::*;
use ToTask::*;
use mdns::Event::*;
use networking::swarm::BehaviourEvent::*;
log::info!("RUST: networking task started");
loop {
tokio::select! {
message = to_task_rx.recv() => {
// handle closed channel
let Some(message) = message else {
log::info!("RUST: channel closed");
break;
};
// dispatch incoming messages
match message {
GossipsubSubscribe { topic, result_tx } => {
// try to subscribe
let result = swarm.behaviour_mut()
.gossipsub.subscribe(&IdentTopic::new(topic));
// send response oneshot
if let Err(e) = result_tx.send(result.pyerr()) {
log::error!("RUST: could not subscribe to gossipsub topic since channel already closed: {e:?}");
continue;
}
}
GossipsubUnsubscribe { topic, result_tx } => {
// try to unsubscribe from the topic
let result = swarm.behaviour_mut()
.gossipsub.unsubscribe(&IdentTopic::new(topic));
// send response oneshot (or exit if connection closed)
if let Err(e) = result_tx.send(result) {
log::error!("RUST: could not unsubscribe from gossipsub topic since channel already closed: {e:?}");
continue;
}
}
GossipsubPublish { topic, data, result_tx } => {
// try to publish the data -> catch NoPeersSubscribedToTopic error & convert to correct exception
let result = swarm.behaviour_mut().gossipsub.publish(
IdentTopic::new(topic), data);
let pyresult: PyResult<MessageId> = if let Err(PublishError::NoPeersSubscribedToTopic) = result {
Err(exception::PyNoPeersSubscribedToTopicError::new_err())
} else if let Err(PublishError::AllQueuesFull(_)) = result {
Err(exception::PyAllQueuesFullError::new_err())
} else {
result.pyerr()
};
// send response oneshot (or exit if connection closed)
if let Err(e) = result_tx.send(pyresult) {
log::error!("RUST: could not publish gossipsub message since channel already closed: {e:?}");
continue;
}
}
}
}
// architectural solution to this problem:
// create keep_alive behavior who's job it is to dial peers discovered by mDNS (and drop when expired)
// -> it will emmit TRUE connected/disconnected events consumable elsewhere
//
// gossipsub will feed off-of dial attempts created by networking, and that will bootstrap its' peers list
// then for actual communication it will dial those peers if need-be
swarm_event = swarm.select_next_some() => {
match swarm_event {
Behaviour(Gossipsub(gossipsub::Event::Message {
message: Message {
topic,
data,
..
},
..
})) => {
// topic-ID is just the topic hash!!! (since we used identity hasher)
let message = (topic.into_string(), data);
// send incoming message to channel (or exit if connection closed)
if let Err(e) = gossipsub_message_tx.send(message).await {
log::error!("RUST: could not send incoming gossipsub message since channel already closed: {e}");
continue;
}
},
Behaviour(Discovery(discovery::Event::ConnectionEstablished { peer_id, remote_ip, remote_tcp_port, .. })) => {
// grab IPv4 string
let remote_ipv4 = match remote_ip {
IpAddr::V4(ip) => ip.to_string(),
IpAddr::V6(ip) => {
log::warn!("RUST: ignoring connection to IPv6 address: {ip}");
continue;
}
};
// send connection event to channel (or exit if connection closed)
if let Err(e) = connection_update_tx.send(PyConnectionUpdate {
update_type: PyConnectionUpdateType::Connected,
peer_id: PyPeerId(peer_id),
remote_ipv4,
remote_tcp_port,
}).await {
log::error!("RUST: could not send connection update since channel already closed: {e}");
continue;
}
},
Behaviour(Discovery(discovery::Event::ConnectionClosed { peer_id, remote_ip, remote_tcp_port, .. })) => {
// grab IPv4 string
let remote_ipv4 = match remote_ip {
IpAddr::V4(ip) => ip.to_string(),
IpAddr::V6(ip) => {
log::warn!("RUST: ignoring disconnection from IPv6 address: {ip}");
continue;
}
};
// send disconnection event to channel (or exit if connection closed)
if let Err(e) = connection_update_tx.send(PyConnectionUpdate {
update_type: PyConnectionUpdateType::Disconnected,
peer_id: PyPeerId(peer_id),
remote_ipv4,
remote_tcp_port,
}).await {
log::error!("RUST: could not send connection update since channel already closed: {e}");
continue;
}
},
e => {
log::info!("RUST: other event {e:?}");
}
}
}
}
}
log::info!("RUST: networking task stopped");
}
#[gen_stub_pyclass]
#[pyclass(name = "NetworkingHandle")]
#[derive(Debug)]
struct PyNetworkingHandle {
// channels
to_task_tx: Option<mpsc::Sender<ToTask>>,
connection_update_rx: Mutex<mpsc::Receiver<PyConnectionUpdate>>,
gossipsub_message_rx: Mutex<mpsc::Receiver<(String, Vec<u8>)>>,
pub to_swarm: mpsc::Sender<ToSwarm>,
pub swarm: Mutex<Swarm>,
}
impl Drop for PyNetworkingHandle {
fn drop(&mut self) {
// TODO: may or may not need to await a "kill-signal" oneshot channel message,
// to ensure that the networking task is done BEFORE exiting the clear function...
// but this may require GIL?? and it may not be safe to call GIL here??
self.to_task_tx = None; // Using Option<T> as a trick to force channel to be dropped
}
#[gen_stub_pyclass_complex_enum]
#[pyclass]
enum PyFromSwarm {
Connection {
peer_id: String,
connected: bool,
},
Message {
origin: String,
topic: String,
data: Py<PyBytes>,
},
}
#[allow(clippy::expect_used)]
impl PyNetworkingHandle {
fn new(
to_task_tx: mpsc::Sender<ToTask>,
connection_update_rx: mpsc::Receiver<PyConnectionUpdate>,
gossipsub_message_rx: mpsc::Receiver<(String, Vec<u8>)>,
) -> Self {
Self {
to_task_tx: Some(to_task_tx),
connection_update_rx: Mutex::new(connection_update_rx),
gossipsub_message_rx: Mutex::new(gossipsub_message_rx),
impl From<FromSwarm> for PyFromSwarm {
fn from(value: FromSwarm) -> Self {
match value {
FromSwarm::Discovered { peer_id } => Self::Connection {
peer_id: peer_id.to_base58(),
connected: true,
},
FromSwarm::Expired { peer_id } => Self::Connection {
peer_id: peer_id.to_base58(),
connected: false,
},
FromSwarm::Message { from, topic, data } => Self::Message {
origin: from.to_base58(),
topic: topic,
data: data.pybytes(),
},
}
}
const fn to_task_tx(&self) -> &mpsc::Sender<ToTask> {
self.to_task_tx
.as_ref()
.expect("The sender should only be None after de-initialization.")
}
}
#[gen_stub_pymethods]
@@ -343,97 +175,32 @@ impl PyNetworkingHandle {
#[new]
fn py_new(identity: Bound<'_, PyKeypair>) -> PyResult<Self> {
use pyo3_async_runtimes::tokio::get_runtime;
// create communication channels
let (to_task_tx, to_task_rx) = mpsc::channel(MPSC_CHANNEL_SIZE);
let (connection_update_tx, connection_update_rx) = mpsc::channel(MPSC_CHANNEL_SIZE);
let (gossipsub_message_tx, gossipsub_message_rx) = mpsc::channel(MPSC_CHANNEL_SIZE);
let (to_swarm, from_client) = mpsc::channel(MPSC_CHANNEL_SIZE);
// get identity
let identity = identity.borrow().0.clone();
// create networking swarm (within tokio context!! or it crashes)
let swarm = get_runtime()
.block_on(async { create_swarm(identity) })
.pyerr()?;
let swarm = { create_swarm(identity, from_client).pyerr()? };
// spawn tokio task running the networking logic
get_runtime().spawn(async move {
networking_task(
swarm,
to_task_rx,
connection_update_tx,
gossipsub_message_tx,
)
.await;
});
Ok(Self::new(
to_task_tx,
connection_update_rx,
gossipsub_message_rx,
))
Ok(Self {
swarm: Mutex::new(swarm),
to_swarm,
})
}
#[gen_stub(skip)]
const fn __traverse__(&self, _visit: PyVisit<'_>) -> Result<(), PyTraverseError> {
Ok(()) // This is needed purely so `__clear__` can work
}
#[gen_stub(skip)]
fn __clear__(&mut self) {
// TODO: may or may not need to await a "kill-signal" oneshot channel message,
// to ensure that the networking task is done BEFORE exiting the clear function...
// but this may require GIL?? and it may not be safe to call GIL here??
self.to_task_tx = None; // Using Option<T> as a trick to force channel to be dropped
}
// ---- Connection update receiver methods ----
/// Receives the next `ConnectionUpdate` from networking.
async fn connection_update_recv(&self) -> PyResult<PyConnectionUpdate> {
self.connection_update_rx
.lock()
.allow_threads_py() // allow-threads-aware async call
.await
.recv_py()
.allow_threads_py() // allow-threads-aware async call
async fn recv(&self) -> PyResult<PyFromSwarm> {
self.swarm
.try_lock()
.expect("tried to recv from swarm twice concurrently")
.next()
.allow_threads_py()
.await
.ok_or(PyErr::receiver_channel_closed())
.map(Into::into)
}
/// 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 fn connection_update_recv_many(&self, limit: usize) -> PyResult<Vec<PyConnectionUpdate>> {
self.connection_update_rx
.lock()
.allow_threads_py() // allow-threads-aware async call
.await
.recv_many_py(limit)
.allow_threads_py() // allow-threads-aware async call
.await
}
// TODO: rn this blocks main thread if anything else is awaiting the channel (bc its a mutex)
// so its too dangerous to expose just yet. figure out a better semantics for handling this,
// so things don't randomly block
// /// Tries to receive the next `ConnectionUpdate` from networking.
// fn connection_update_try_recv(&self) -> PyResult<Option<PyConnectionUpdate>> {
// self.connection_update_rx.blocking_lock().try_recv_py()
// }
//
// /// Checks if the `ConnectionUpdate` channel is empty.
// fn connection_update_is_empty(&self) -> bool {
// self.connection_update_rx.blocking_lock().is_empty()
// }
//
// /// Returns the number of `ConnectionUpdate`s in the channel.
// fn connection_update_len(&self) -> usize {
// self.connection_update_rx.blocking_lock().len()
// }
// ---- Gossipsub management methods ----
/// Subscribe to a `GossipSub` topic.
@@ -443,10 +210,10 @@ impl PyNetworkingHandle {
let (tx, rx) = oneshot::channel();
// send off request to subscribe
self.to_task_tx()
.send_py(ToTask::GossipsubSubscribe {
self.to_swarm
.send_py(ToSwarm::Subscribe {
topic,
result_tx: tx,
result_sender: tx,
})
.allow_threads_py() // allow-threads-aware async call
.await?;
@@ -455,6 +222,7 @@ impl PyNetworkingHandle {
rx.allow_threads_py() // allow-threads-aware async call
.await
.map_err(|_| PyErr::receiver_channel_closed())?
.pyerr()
}
/// Unsubscribes from a `GossipSub` topic.
@@ -464,10 +232,10 @@ impl PyNetworkingHandle {
let (tx, rx) = oneshot::channel();
// send off request to unsubscribe
self.to_task_tx()
.send_py(ToTask::GossipsubUnsubscribe {
self.to_swarm
.send_py(ToSwarm::Unsubscribe {
topic,
result_tx: tx,
result_sender: tx,
})
.allow_threads_py() // allow-threads-aware async call
.await?;
@@ -485,12 +253,12 @@ impl PyNetworkingHandle {
let (tx, rx) = oneshot::channel();
// send off request to subscribe
let data = Python::with_gil(|py| Vec::from(data.as_bytes(py)));
self.to_task_tx()
.send_py(ToTask::GossipsubPublish {
let data = Python::attach(|py| Vec::from(data.as_bytes(py)));
self.to_swarm
.send_py(ToSwarm::Publish {
topic,
data,
result_tx: tx,
result_sender: tx,
})
.allow_threads_py() // allow-threads-aware async call
.await?;
@@ -499,64 +267,14 @@ impl PyNetworkingHandle {
let _ = rx
.allow_threads_py() // allow-threads-aware async call
.await
.map_err(|_| PyErr::receiver_channel_closed())??;
.map_err(|_| PyErr::receiver_channel_closed())?
.map_err(|e| match e {
PublishError::AllQueuesFull(_) => PyAllQueuesFullError::new_err(),
PublishError::MessageTooLarge => PyNoPeersSubscribedToTopicError::new_err(),
e => PyRuntimeError::new_err(e.to_string()),
})?;
Ok(())
}
// ---- Gossipsub message receiver methods ----
/// Receives the next message from the `GossipSub` network.
async fn gossipsub_recv(&self) -> PyResult<(String, Py<PyBytes>)> {
self.gossipsub_message_rx
.lock()
.allow_threads_py() // allow-threads-aware async call
.await
.recv_py()
.allow_threads_py() // allow-threads-aware async call
.await
.map(|(t, d)| (t, d.pybytes()))
}
/// 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.
async fn gossipsub_recv_many(&self, limit: usize) -> PyResult<Vec<(String, Py<PyBytes>)>> {
Ok(self
.gossipsub_message_rx
.lock()
.allow_threads_py() // allow-threads-aware async call
.await
.recv_many_py(limit)
.allow_threads_py() // allow-threads-aware async call
.await?
.into_iter()
.map(|(t, d)| (t, d.pybytes()))
.collect())
}
// TODO: rn this blocks main thread if anything else is awaiting the channel (bc its a mutex)
// so its too dangerous to expose just yet. figure out a better semantics for handling this,
// so things don't randomly block
// /// Tries to receive the next message from the `GossipSub` network.
// fn gossipsub_try_recv(&self) -> PyResult<Option<(String, Py<PyBytes>)>> {
// Ok(self
// .gossipsub_message_rx
// .blocking_lock()
// .try_recv_py()?
// .map(|(t, d)| (t, d.pybytes())))
// }
//
// /// Checks if the `GossipSub` message channel is empty.
// fn gossipsub_is_empty(&self) -> bool {
// self.gossipsub_message_rx.blocking_lock().is_empty()
// }
//
// /// Returns the number of `GossipSub` messages in the channel.
// fn gossipsub_len(&self) -> usize {
// self.gossipsub_message_rx.blocking_lock().len()
// }
}
pub fn networking_submodule(m: &Bound<'_, PyModule>) -> PyResult<()> {

View File

@@ -1,159 +0,0 @@
use crate::ext::ResultExt as _;
use libp2p::PeerId;
use libp2p::identity::Keypair;
use pyo3::prelude::{PyBytesMethods as _, PyModule, PyModuleMethods as _};
use pyo3::types::PyBytes;
use pyo3::{Bound, PyResult, Python, pyclass, pymethods};
use pyo3_stub_gen::derive::{gen_stub_pyclass, gen_stub_pymethods};
/// Identity keypair of a node.
#[gen_stub_pyclass]
#[pyclass(name = "Keypair", frozen)]
#[repr(transparent)]
pub struct PyKeypair(pub Keypair);
#[gen_stub_pymethods]
#[pymethods]
#[allow(clippy::needless_pass_by_value)]
impl PyKeypair {
/// Generate a new Ed25519 keypair.
#[staticmethod]
fn generate_ed25519() -> Self {
Self(Keypair::generate_ed25519())
}
/// Generate a new ECDSA keypair.
#[staticmethod]
fn generate_ecdsa() -> Self {
Self(Keypair::generate_ecdsa())
}
/// Generate a new Secp256k1 keypair.
#[staticmethod]
fn generate_secp256k1() -> Self {
Self(Keypair::generate_secp256k1())
}
/// Decode a private key from a protobuf structure and parse it as a `Keypair`.
#[staticmethod]
fn from_protobuf_encoding(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
let bytes = Vec::from(bytes.as_bytes());
Ok(Self(Keypair::from_protobuf_encoding(&bytes).pyerr()?))
}
/// 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]
fn rsa_from_pkcs8(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
let mut bytes = Vec::from(bytes.as_bytes());
Ok(Self(Keypair::rsa_from_pkcs8(&mut bytes).pyerr()?))
}
/// 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]
fn secp256k1_from_der(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
let mut bytes = Vec::from(bytes.as_bytes());
Ok(Self(Keypair::secp256k1_from_der(&mut bytes).pyerr()?))
}
#[staticmethod]
fn ed25519_from_bytes(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
let mut bytes = Vec::from(bytes.as_bytes());
Ok(Self(Keypair::ed25519_from_bytes(&mut bytes).pyerr()?))
}
/// Encode a private key as protobuf structure.
fn to_protobuf_encoding<'py>(&self, py: Python<'py>) -> PyResult<Bound<'py, PyBytes>> {
let bytes = self.0.to_protobuf_encoding().pyerr()?;
Ok(PyBytes::new(py, &bytes))
}
/// Convert the `Keypair` into the corresponding `PeerId`.
fn to_peer_id(&self) -> PyPeerId {
PyPeerId(self.0.public().to_peer_id())
}
// /// Hidden constructor for pickling support. TODO: figure out how to do pickling...
// #[gen_stub(skip)]
// #[new]
// fn py_new(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
// Self::from_protobuf_encoding(bytes)
// }
//
// #[gen_stub(skip)]
// fn __setstate__(&mut self, state: Bound<'_, PyBytes>) -> PyResult<()> {
// *self = Self::from_protobuf_encoding(state)?;
// Ok(())
// }
//
// #[gen_stub(skip)]
// fn __getstate__<'py>(&self, py: Python<'py>) -> PyResult<Bound<'py, PyBytes>> {
// self.to_protobuf_encoding(py)
// }
//
// #[gen_stub(skip)]
// pub fn __getnewargs__<'py>(&self, py: Python<'py>) -> PyResult<(Bound<'py, PyBytes>,)> {
// Ok((self.to_protobuf_encoding(py)?,))
// }
}
/// 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).
#[gen_stub_pyclass]
#[pyclass(name = "PeerId", frozen)]
#[derive(Debug, Clone)]
#[repr(transparent)]
pub struct PyPeerId(pub PeerId);
#[gen_stub_pymethods]
#[pymethods]
#[allow(clippy::needless_pass_by_value)]
impl PyPeerId {
/// Generates a random peer ID from a cryptographically secure PRNG.
///
/// This is useful for randomly walking on a DHT, or for testing purposes.
#[staticmethod]
fn random() -> Self {
Self(PeerId::random())
}
/// Parses a `PeerId` from bytes.
#[staticmethod]
fn from_bytes(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
let bytes = Vec::from(bytes.as_bytes());
Ok(Self(PeerId::from_bytes(&bytes).pyerr()?))
}
/// Returns a raw bytes representation of this `PeerId`.
fn to_bytes<'py>(&self, py: Python<'py>) -> Bound<'py, PyBytes> {
let bytes = self.0.to_bytes();
PyBytes::new(py, &bytes)
}
/// Returns a base-58 encoded string of this `PeerId`.
fn to_base58(&self) -> String {
self.0.to_base58()
}
fn __repr__(&self) -> String {
format!("PeerId({})", self.to_base58())
}
fn __str__(&self) -> String {
self.to_base58()
}
}
pub fn ident_submodule(m: &Bound<'_, PyModule>) -> PyResult<()> {
m.add_class::<PyKeypair>()?;
m.add_class::<PyPeerId>()?;
Ok(())
}

View File

@@ -1,8 +0,0 @@
//! A module for exposing Rust's libp2p datatypes over Pyo3
//!
//! TODO: right now we are coupled to libp2p's identity, but eventually we want to create our own
//! independent identity type of some kind or another. This may require handshaking.
//!
pub mod ident;
pub mod multiaddr;

View File

@@ -1,81 +0,0 @@
use crate::ext::ResultExt as _;
use libp2p::Multiaddr;
use pyo3::prelude::{PyBytesMethods as _, PyModule, PyModuleMethods as _};
use pyo3::types::PyBytes;
use pyo3::{Bound, PyResult, Python, pyclass, pymethods};
use pyo3_stub_gen::derive::{gen_stub_pyclass, gen_stub_pymethods};
use std::str::FromStr as _;
/// Representation of a Multiaddr.
#[gen_stub_pyclass]
#[pyclass(name = "Multiaddr", frozen)]
#[derive(Debug, Clone)]
#[repr(transparent)]
pub struct PyMultiaddr(pub Multiaddr);
#[gen_stub_pymethods]
#[pymethods]
#[allow(clippy::needless_pass_by_value)]
impl PyMultiaddr {
/// Create a new, empty multiaddress.
#[staticmethod]
fn empty() -> Self {
Self(Multiaddr::empty())
}
/// Create a new, empty multiaddress with the given capacity.
#[staticmethod]
fn with_capacity(n: usize) -> Self {
Self(Multiaddr::with_capacity(n))
}
/// Parse a `Multiaddr` value from its byte slice representation.
#[staticmethod]
fn from_bytes(bytes: Bound<'_, PyBytes>) -> PyResult<Self> {
let bytes = Vec::from(bytes.as_bytes());
Ok(Self(Multiaddr::try_from(bytes).pyerr()?))
}
/// Parse a `Multiaddr` value from its string representation.
#[staticmethod]
fn from_string(string: String) -> PyResult<Self> {
Ok(Self(Multiaddr::from_str(&string).pyerr()?))
}
/// Return the length in bytes of this multiaddress.
fn len(&self) -> usize {
self.0.len()
}
/// Returns true if the length of this multiaddress is 0.
fn is_empty(&self) -> bool {
self.0.is_empty()
}
/// Return a copy of this [`Multiaddr`]'s byte representation.
fn to_bytes<'py>(&self, py: Python<'py>) -> Bound<'py, PyBytes> {
let bytes = self.0.to_vec();
PyBytes::new(py, &bytes)
}
/// Convert a Multiaddr to a string.
fn to_string(&self) -> String {
self.0.to_string()
}
#[gen_stub(skip)]
fn __repr__(&self) -> String {
format!("Multiaddr({})", self.0)
}
#[gen_stub(skip)]
fn __str__(&self) -> String {
self.to_string()
}
}
pub fn multiaddr_submodule(m: &Bound<'_, PyModule>) -> PyResult<()> {
m.add_class::<PyMultiaddr>()?;
Ok(())
}

View File

@@ -19,21 +19,14 @@ either = { workspace = true }
# macro dependencies
extend = { workspace = true }
delegate = { workspace = true }
impl-trait-for-tuples = { workspace = true }
derive_more = { workspace = true }
# async
tokio = { workspace = true, features = ["full"] }
futures = { workspace = true }
futures-lite = { workspace = true }
futures-timer = { workspace = true }
# utility dependencies
util = { workspace = true }
thiserror = { workspace = true }
#internment = { workspace = true }
#recursion = { workspace = true }
#generativity = { workspace = true }
#itertools = { workspace = true }
tracing-subscriber = { version = "0.3.19", features = ["default", "env-filter"] }
keccak-const = { workspace = true }
@@ -41,4 +34,5 @@ keccak-const = { workspace = true }
log = { workspace = true }
# networking
libp2p = { workspace = true, features = ["full"] }
libp2p = { workspace = true, features = ["full"] }
pin-project = "1.1.10"

View File

@@ -1,6 +1,8 @@
use futures::stream::StreamExt as _;
use libp2p::{gossipsub, identity, swarm::SwarmEvent};
use networking::{discovery, swarm};
use futures_lite::StreamExt;
use libp2p::identity;
use networking::swarm;
use networking::swarm::{FromSwarm, ToSwarm};
use tokio::sync::{mpsc, oneshot};
use tokio::{io, io::AsyncBufReadExt as _, select};
use tracing_subscriber::EnvFilter;
use tracing_subscriber::filter::LevelFilter;
@@ -11,17 +13,23 @@ async fn main() {
.with_env_filter(EnvFilter::from_default_env().add_directive(LevelFilter::INFO.into()))
.try_init();
let (to_swarm, from_client) = mpsc::channel(20);
// Configure swarm
let mut swarm =
swarm::create_swarm(identity::Keypair::generate_ed25519()).expect("Swarm creation failed");
let mut swarm = swarm::create_swarm(identity::Keypair::generate_ed25519(), from_client)
.expect("Swarm creation failed");
// Create a Gossipsub topic & subscribe
let topic = gossipsub::IdentTopic::new("test-net");
swarm
.behaviour_mut()
.gossipsub
.subscribe(&topic)
.expect("Subscribing to topic failed");
let (tx, rx) = oneshot::channel();
_ = to_swarm
.send(ToSwarm::Subscribe {
topic: "test-net".to_string(),
result_sender: tx,
})
.await
.expect("should send");
let mut fused = futures_lite::future::fuse(rx);
// Read full lines from stdin
let mut stdin = io::BufReader::new(io::stdin()).lines();
@@ -32,43 +40,23 @@ async fn main() {
select! {
// on gossipsub outgoing
Ok(Some(line)) = stdin.next_line() => {
if let Err(e) = swarm
.behaviour_mut().gossipsub
.publish(topic.clone(), line.as_bytes()) {
let (tx, rx) = oneshot::channel();
if let Err(e) = to_swarm.send(swarm::ToSwarm::Publish { topic: "test-net".to_string(), data: line.as_bytes().to_vec(), result_sender: tx }).await {
println!("Send error: {e:?}");
return
};
if let Err(e) = rx.await {
println!("Publish error: {e:?}");
}
}
event = swarm.select_next_some() => match event {
},
event = swarm.next() => match event {
// on gossipsub incoming
SwarmEvent::Behaviour(swarm::BehaviourEvent::Gossipsub(gossipsub::Event::Message {
propagation_source: peer_id,
message_id: id,
message,
})) => println!(
"\n\nGot message: '{}' with id: {id} from peer: {peer_id}\n\n",
String::from_utf8_lossy(&message.data),
),
// on discovery
SwarmEvent::Behaviour(swarm::BehaviourEvent::Discovery(e)) => match e {
discovery::Event::ConnectionEstablished {
peer_id, connection_id, remote_ip, remote_tcp_port
} => {
println!("\n\nConnected to: {peer_id}; connection ID: {connection_id}; remote IP: {remote_ip}; remote TCP port: {remote_tcp_port}\n\n");
}
discovery::Event::ConnectionClosed {
peer_id, connection_id, remote_ip, remote_tcp_port
} => {
eprintln!("\n\nDisconnected from: {peer_id}; connection ID: {connection_id}; remote IP: {remote_ip}; remote TCP port: {remote_tcp_port}\n\n");
}
}
// ignore outgoing errors: those are normal
e@SwarmEvent::OutgoingConnectionError { .. } => { log::debug!("Outgoing connection error: {e:?}"); }
// otherwise log any other event
e => { log::info!("Other event {e:?}"); }
}
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::Message { from, topic, data }) => { println!("{topic}/{from}:\n{}", String::from_utf8_lossy(&data)) },
None => {},
},
f = &mut fused => {assert!(f.expect("should recv").expect("should subscribe"))},
}
}
}

View File

@@ -1,127 +0,0 @@
// Copyright 2018 Parity Technologies (UK) Ltd.
//
// Permission is hereby granted, free of charge, to any person obtaining a
// copy of this software and associated documentation files (the "Software"),
// to deal in the Software without restriction, including without limitation
// the rights to use, copy, modify, merge, publish, distribute, sublicense,
// and/or sell copies of the Software, and to permit persons to whom the
// Software is furnished to do so, subject to the following conditions:
//
// The above copyright notice and this permission notice shall be included in
// all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
// FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
// DEALINGS IN THE SOFTWARE.
use futures::stream::StreamExt;
use libp2p::{
gossipsub, mdns, noise,
swarm::{NetworkBehaviour, SwarmEvent},
tcp, yamux,
};
use std::time::Duration;
use std::{error::Error, hash::Hash};
use tokio::{io, io::AsyncBufReadExt, select};
use tracing_subscriber::EnvFilter;
// We create a custom network behaviour that combines Gossipsub and Mdns.
#[derive(NetworkBehaviour)]
struct MyBehaviour {
gossipsub: gossipsub::Behaviour,
mdns: mdns::tokio::Behaviour,
}
#[tokio::main]
async fn main() -> Result<(), Box<dyn Error>> {
let _ = tracing_subscriber::fmt()
.with_env_filter(EnvFilter::from_default_env())
.try_init();
let mut swarm = libp2p::SwarmBuilder::with_new_identity()
.with_tokio()
.with_tcp(
tcp::Config::default(),
noise::Config::new,
yamux::Config::default,
)?
.with_behaviour(|key| {
// Set a custom gossipsub configuration
let gossipsub_config = gossipsub::ConfigBuilder::default()
.heartbeat_interval(Duration::from_secs(10))
.validation_mode(gossipsub::ValidationMode::Strict) // This sets the kind of message validation. The default is Strict (enforce message signing)
.build()
.map_err(io::Error::other)?; // Temporary hack because `build` does not return a proper `std::error::Error`.
// build a gossipsub network behaviour
let gossipsub = gossipsub::Behaviour::new(
gossipsub::MessageAuthenticity::Signed(key.clone()),
gossipsub_config,
)?;
let mdns =
mdns::tokio::Behaviour::new(mdns::Config::default(), key.public().to_peer_id())?;
Ok(MyBehaviour { gossipsub, mdns })
})?
.build();
println!("Running swarm with identity {}", swarm.local_peer_id());
// Create a Gossipsub topic
let topic = gossipsub::IdentTopic::new("test-net");
// subscribes to our topic
swarm.behaviour_mut().gossipsub.subscribe(&topic)?;
// Read full lines from stdin
let mut stdin = io::BufReader::new(io::stdin()).lines();
// Listen on all interfaces and whatever port the OS assigns
swarm.listen_on("/ip4/0.0.0.0/tcp/0".parse()?)?;
println!("Enter messages via STDIN and they will be sent to connected peers using Gossipsub");
// Kick it off
loop {
select! {
Ok(Some(line)) = stdin.next_line() => {
if let Err(e) = swarm
.behaviour_mut().gossipsub
.publish(topic.clone(), line.as_bytes()) {
println!("Publish error: {e:?}");
}
}
event = swarm.select_next_some() => match event {
SwarmEvent::Behaviour(MyBehaviourEvent::Mdns(mdns::Event::Discovered(list))) => {
for (peer_id, multiaddr) in list {
println!("mDNS discovered a new peer: {peer_id} on {multiaddr}");
swarm.behaviour_mut().gossipsub.add_explicit_peer(&peer_id);
}
},
SwarmEvent::Behaviour(MyBehaviourEvent::Mdns(mdns::Event::Expired(list))) => {
for (peer_id, multiaddr) in list {
println!("mDNS discover peer has expired: {peer_id} on {multiaddr}");
swarm.behaviour_mut().gossipsub.remove_explicit_peer(&peer_id);
}
},
SwarmEvent::Behaviour(MyBehaviourEvent::Gossipsub(gossipsub::Event::Message {
propagation_source: peer_id,
message_id: id,
message,
})) => println!(
"Got message: '{}' with id: {id} from peer: {peer_id}",
String::from_utf8_lossy(&message.data),
),
SwarmEvent::NewListenAddr { address, .. } => {
println!("Local node is listening on {address}");
}
e => {
println!("Other swarm event: {:?}", e);
}
}
}
}
}

View File

@@ -1,8 +1,7 @@
use crate::ext::MultiaddrExt;
use crate::keep_alive;
use delegate::delegate;
use either::Either;
use futures::FutureExt;
use futures_lite::FutureExt;
use futures_timer::Delay;
use libp2p::core::transport::PortUse;
use libp2p::core::{ConnectedPoint, Endpoint};
@@ -363,7 +362,7 @@ impl NetworkBehaviour for Behaviour {
}
// retry connecting to all mDNS peers periodically (fails safely if already connected)
if self.retry_delay.poll_unpin(cx).is_ready() {
if self.retry_delay.poll(cx).is_ready() {
for (p, mas) in self.mdns_discovered.clone() {
for ma in mas {
self.dial(p, ma)

View File

@@ -1,44 +0,0 @@
use delegate::delegate;
use libp2p::swarm::handler::ConnectionEvent;
use libp2p::swarm::{ConnectionHandlerEvent, SubstreamProtocol, dummy, handler};
use std::task::{Context, Poll};
/// An implementation of [`ConnectionHandler`] that doesn't handle any protocols, but it keeps
/// the connection alive.
#[derive(Clone)]
#[repr(transparent)]
pub struct ConnectionHandler(dummy::ConnectionHandler);
impl ConnectionHandler {
pub fn new() -> Self {
ConnectionHandler(dummy::ConnectionHandler)
}
}
impl handler::ConnectionHandler for ConnectionHandler {
// delegate types and implementation mostly to dummy handler
type FromBehaviour = <dummy::ConnectionHandler as handler::ConnectionHandler>::FromBehaviour;
type ToBehaviour = <dummy::ConnectionHandler as handler::ConnectionHandler>::ToBehaviour;
type InboundProtocol =
<dummy::ConnectionHandler as handler::ConnectionHandler>::InboundProtocol;
type OutboundProtocol =
<dummy::ConnectionHandler as handler::ConnectionHandler>::OutboundProtocol;
type InboundOpenInfo =
<dummy::ConnectionHandler as handler::ConnectionHandler>::InboundOpenInfo;
type OutboundOpenInfo =
<dummy::ConnectionHandler as handler::ConnectionHandler>::OutboundOpenInfo;
delegate! {
to self.0 {
fn listen_protocol(&self) -> SubstreamProtocol<Self::InboundProtocol, Self::InboundOpenInfo>;
fn poll(&mut self, cx: &mut Context<'_>) -> Poll<ConnectionHandlerEvent<Self::OutboundProtocol, Self::OutboundOpenInfo, Self::ToBehaviour>>;
fn on_behaviour_event(&mut self, event: Self::FromBehaviour);
fn on_connection_event(&mut self, event: ConnectionEvent<Self::InboundProtocol, Self::OutboundProtocol, Self::InboundOpenInfo, Self::OutboundOpenInfo>);
}
}
// specifically override this to force connection to stay alive
fn connection_keep_alive(&self) -> bool {
true
}
}

View File

@@ -3,19 +3,7 @@
//! this is here as a placeholder documentation
//!
//!
// enable Rust-unstable features for convenience
#![feature(trait_alias)]
// #![feature(stmt_expr_attributes)]
// #![feature(unboxed_closures)]
// #![feature(assert_matches)]
// #![feature(async_fn_in_dyn_trait)]
// #![feature(async_for_loop)]
// #![feature(auto_traits)]
// #![feature(negative_impls)]
pub mod discovery;
pub mod keep_alive;
pub mod swarm;
/// Namespace for all the type/trait aliases used by this crate.
@@ -54,11 +42,3 @@ pub(crate) mod ext {
}
}
}
pub(crate) mod private {
#![allow(dead_code)]
/// Sealed traits support
pub trait Sealed {}
impl<T: ?Sized> Sealed for T {}
}

View File

@@ -1,9 +1,11 @@
use crate::alias;
use crate::swarm::transport::tcp_transport;
pub use behaviour::{Behaviour, BehaviourEvent};
use libp2p::{SwarmBuilder, identity};
use std::task::Poll;
pub type Swarm = libp2p::Swarm<Behaviour>;
use crate::swarm::transport::tcp_transport;
use crate::{alias, discovery};
pub use behaviour::{Behaviour, BehaviourEvent};
use futures_lite::Stream;
use libp2p::{PeerId, SwarmBuilder, gossipsub, identity, swarm::SwarmEvent};
use tokio::sync::{mpsc, oneshot};
/// The current version of the network: this prevents devices running different versions of the
/// software from interacting with each other.
@@ -15,8 +17,133 @@ pub type Swarm = libp2p::Swarm<Behaviour>;
pub const NETWORK_VERSION: &[u8] = b"v0.0.1";
pub const OVERRIDE_VERSION_ENV_VAR: &str = "EXO_LIBP2P_NAMESPACE";
pub enum ToSwarm {
Unsubscribe {
topic: String,
result_sender: oneshot::Sender<bool>,
},
Subscribe {
topic: String,
result_sender: oneshot::Sender<Result<bool, gossipsub::SubscriptionError>>,
},
Publish {
topic: String,
data: Vec<u8>,
result_sender: oneshot::Sender<Result<gossipsub::MessageId, gossipsub::PublishError>>,
},
}
pub enum FromSwarm {
Message {
from: PeerId,
topic: String,
data: Vec<u8>,
},
Discovered {
peer_id: PeerId,
},
Expired {
peer_id: PeerId,
},
}
#[pin_project::pin_project]
pub struct Swarm {
#[pin]
inner: libp2p::Swarm<Behaviour>,
from_client: mpsc::Receiver<ToSwarm>,
}
impl Swarm {
fn on_message(&mut self, message: ToSwarm) {
match message {
ToSwarm::Subscribe {
topic,
result_sender,
} => {
// try to subscribe
let result = self
.inner
.behaviour_mut()
.gossipsub
.subscribe(&gossipsub::IdentTopic::new(topic));
// send response oneshot
_ = result_sender.send(result)
}
ToSwarm::Unsubscribe {
topic,
result_sender,
} => {
// try to unsubscribe from the topic
let result = self
.inner
.behaviour_mut()
.gossipsub
.unsubscribe(&gossipsub::IdentTopic::new(topic));
// send response oneshot (or exit if connection closed)
_ = result_sender.send(result)
}
ToSwarm::Publish {
topic,
data,
result_sender,
} => {
// try to publish the data -> catch NoPeersSubscribedToTopic error & convert to correct exception
let result = self
.inner
.behaviour_mut()
.gossipsub
.publish(gossipsub::IdentTopic::new(topic), data);
// send response oneshot (or exit if connection closed)
_ = result_sender.send(result)
}
}
}
}
impl Stream for Swarm {
type Item = FromSwarm;
fn poll_next(
mut self: std::pin::Pin<&mut Self>,
cx: &mut std::task::Context<'_>,
) -> Poll<Option<Self::Item>> {
match self.from_client.poll_recv(cx) {
Poll::Ready(Some(msg)) => self.on_message(msg),
Poll::Ready(None) => return Poll::Ready(None),
Poll::Pending => {}
}
match self.project().inner.poll_next(cx) {
Poll::Pending => Poll::Pending,
Poll::Ready(None) => Poll::Ready(None),
Poll::Ready(Some(swarm_event)) => match swarm_event {
SwarmEvent::Behaviour(BehaviourEvent::Gossipsub(gossipsub::Event::Message {
message:
gossipsub::Message {
source: Some(peer_id),
topic,
data,
..
},
..
})) => Poll::Ready(Some(FromSwarm::Message {
from: peer_id,
topic: topic.into_string(),
data,
})),
SwarmEvent::Behaviour(BehaviourEvent::Discovery(
discovery::Event::ConnectionEstablished { peer_id, .. },
)) => Poll::Ready(Some(FromSwarm::Discovered { peer_id })),
SwarmEvent::Behaviour(BehaviourEvent::Discovery(
discovery::Event::ConnectionClosed { peer_id, .. },
)) => Poll::Ready(Some(FromSwarm::Expired { peer_id })),
_ => Poll::Pending,
},
}
}
}
/// Create and configure a swarm which listens to all ports on OS
pub fn create_swarm(keypair: identity::Keypair) -> alias::AnyResult<Swarm> {
pub fn create_swarm(
keypair: identity::Keypair,
from_client: mpsc::Receiver<ToSwarm>,
) -> alias::AnyResult<Swarm> {
let mut swarm = SwarmBuilder::with_existing_identity(keypair)
.with_tokio()
.with_other_transport(tcp_transport)?
@@ -25,13 +152,16 @@ pub fn create_swarm(keypair: identity::Keypair) -> alias::AnyResult<Swarm> {
// Listen on all interfaces and whatever port the OS assigns
swarm.listen_on("/ip4/0.0.0.0/tcp/0".parse()?)?;
Ok(swarm)
Ok(Swarm {
inner: swarm,
from_client,
})
}
mod transport {
use crate::alias;
use crate::swarm::{NETWORK_VERSION, OVERRIDE_VERSION_ENV_VAR};
use futures::{AsyncRead, AsyncWrite};
use futures_lite::{AsyncRead, AsyncWrite};
use keccak_const::Sha3_256;
use libp2p::core::muxing;
use libp2p::core::transport::Boxed;

View File

@@ -1,11 +1,10 @@
{ inputs, ... }:
{
perSystem =
{ config, self', inputs', pkgs, lib, ... }:
{ inputs', pkgs, lib, ... }:
let
# Fenix nightly toolchain with all components
fenixPkgs = inputs'.fenix.packages;
rustToolchain = fenixPkgs.complete.withComponents [
rustToolchain = inputs'.fenix.packages.stable.withComponents [
"cargo"
"rustc"
"clippy"

View File

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

View File

@@ -47,6 +47,7 @@ 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)
@@ -62,6 +63,8 @@ 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:
@@ -107,13 +110,17 @@ class DownloadCoordinator:
self._last_progress_time[model_id] = current_time()
async def run(self) -> None:
logger.info("Starting DownloadCoordinator")
self._test_internet_connection()
logger.info(
f"Starting DownloadCoordinator{' (offline mode)' if self.offline else ''}"
)
if not self.offline:
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)
tg.start_soon(self._check_internet_connection)
if not self.offline:
tg.start_soon(self._check_internet_connection)
def _test_internet_connection(self) -> None:
try:
@@ -202,6 +209,20 @@ 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)

View File

@@ -448,12 +448,13 @@ 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
model_id, revision, path, target_dir, on_progress, skip_internet
)
except HuggingFaceAuthenticationError:
raise
@@ -487,10 +488,14 @@ 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
@@ -510,6 +515,11 @@ 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
@@ -814,6 +824,7 @@ 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:

View File

@@ -0,0 +1,230 @@
"""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)

View File

@@ -39,12 +39,13 @@ 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_node_id())
session_id = SessionId(master_node_id=node_id, election_clock=0)
router = Router.create(keypair)
await router.register_topic(topics.GLOBAL_EVENTS)
@@ -68,6 +69,7 @@ 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
@@ -132,6 +134,7 @@ class Node:
api,
node_id,
event_index_counter,
args.offline,
)
async def run(self):
@@ -222,6 +225,7 @@ 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:
@@ -260,6 +264,9 @@ def main():
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")
# Set FAST_SYNCH override env var for runner subprocesses
if args.fast_synch is True:
os.environ["EXO_FAST_SYNCH"] = "on"
@@ -282,6 +289,7 @@ class Args(CamelCaseModel):
tb_only: bool = False
no_worker: bool = False
no_downloads: bool = False
offline: bool = False
fast_synch: bool | None = None # None = auto, True = force on, False = force off
@classmethod
@@ -329,6 +337,11 @@ class Args(CamelCaseModel):
action="store_true",
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",
)
fast_synch_group = parser.add_mutually_exclusive_group()
fast_synch_group.add_argument(
"--fast-synch",

View File

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

View File

@@ -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_node_id())
session_id = SessionId(master_node_id=node_id, election_clock=0)
ge_sender, global_event_receiver = channel[ForwarderEvent]()

View File

@@ -30,7 +30,7 @@ class ConnectionMessage(CamelCaseModel):
@classmethod
def from_update(cls, update: ConnectionUpdate) -> "ConnectionMessage":
return cls(
node_id=NodeId(update.peer_id.to_base58()),
node_id=NodeId(update.peer_id),
connection_type=ConnectionMessageType.from_update_type(update.update_type),
remote_ipv4=update.remote_ipv4,
remote_tcp_port=update.remote_tcp_port,

View File

@@ -221,7 +221,7 @@ 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")
@@ -235,12 +235,12 @@ def get_node_id_keypair(
protobuf_encoded = f.read()
try: # if decoded successfully, save & return
return Keypair.from_protobuf_encoding(protobuf_encoded)
return Keypair.from_bytes(protobuf_encoded)
except ValueError as e: # on runtime error, assume corrupt file
logger.warning(f"Encountered error when trying to get keypair: {e}")
# if no valid credentials, create new ones and persist
with open(path, "w+b") as f:
keypair = Keypair.generate_ed25519()
f.write(keypair.to_protobuf_encoding())
f.write(keypair.to_bytes())
return keypair

View File

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

View File

@@ -23,7 +23,7 @@ def _get_keypair_concurrent_subprocess_task(
sem.release()
# wait to be told to begin simultaneous read
ev.wait()
queue.put(get_node_id_keypair().to_protobuf_encoding())
queue.put(get_node_id_keypair().to_bytes())
def _get_keypair_concurrent(num_procs: int) -> bytes:

View File

@@ -1,9 +1,9 @@
import time
from collections.abc import Generator
from typing import Annotated, Any, Literal
from typing import Annotated, Any, Literal, get_args
from uuid import uuid4
from pydantic import BaseModel, Field
from pydantic import BaseModel, Field, field_validator
from exo.shared.models.model_cards import ModelCard, ModelId
from exo.shared.types.common import CommandId, NodeId
@@ -262,6 +262,27 @@ 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
@@ -281,7 +302,7 @@ 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: str | None = "1024x1024"
size: ImageSize = "auto"
stream: bool | None = False
style: str | None = "vivid"
user: str | None = None
@@ -289,6 +310,11 @@ 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
@@ -305,13 +331,18 @@ 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: str | None = "1024x1024"
size: ImageSize = "auto"
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":

View File

@@ -14,6 +14,7 @@ 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 (
@@ -23,9 +24,9 @@ from exo.shared.types.worker.runner_response import (
from exo.worker.engines.image.distributed_model import DistributedImageModel
def parse_size(size_str: str | None) -> tuple[int, int]:
def parse_size(size_str: ImageSize) -> tuple[int, int]:
"""Parse size parameter like '1024x1024' to (width, height) tuple."""
if not size_str:
if size_str == "auto":
return (1024, 1024)
try:
@@ -109,6 +110,9 @@ 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

View File

@@ -163,11 +163,14 @@ class PipelineLastLayer(CustomMlxLayer):
output, (self.r + 1) % self.s, group=self.group
)
if cache is not None:
cache.keys = mx.depends(cache.keys, output) # type: ignore[reportUnknownMemberType]
# 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
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)[
@@ -307,7 +310,9 @@ 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:
cache[-1].state = mx.depends(cache[-1].state, logits) # type: ignore
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
return logits
@@ -333,7 +338,9 @@ 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]
cache[-1].state = mx.depends(cache[-1].state, logits) # pyright: ignore[reportAny,reportUnknownMemberType]
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]
return logits
@@ -547,10 +554,12 @@ 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(
@@ -581,12 +590,18 @@ 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 in place since the MoE should be responsible
# for aggregating the results.
# Shard the MoE.
else:
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)
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.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)
@@ -779,8 +794,7 @@ class MiniMaxShardingStrategy(TensorParallelShardingStrategy):
layer.self_attn = WrappedMiniMaxAttention(layer.self_attn, self.group) # pyright: ignore[reportAttributeAccessIssue,reportArgumentType]
# Shard the MoE. Shard in place since the MoE should be responsible
# for aggregating the results.
# Shard the MoE.
self.all_to_sharded_linear_in_place(
layer.block_sparse_moe.switch_mlp.gate_proj
)
@@ -893,8 +907,7 @@ class QwenShardingStrategy(TensorParallelShardingStrategy):
layer.self_attn.num_attention_heads //= self.N
layer.self_attn.num_key_value_heads //= self.N
# Shard the MoE. Shard in place since the MoE should be responsible
# for aggregating the results.
# Shard the MoE.
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)

View File

@@ -57,6 +57,7 @@ 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.
@@ -86,6 +87,9 @@ 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(
@@ -305,16 +309,9 @@ 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,
model, tokenizer, sampler, prompt_tokens[:-1], caches, group
)
cache_snapshots: list[CacheSnapshot] | None = ssm_snapshots_list or None
@@ -331,6 +328,7 @@ 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(

View File

@@ -285,10 +285,12 @@ 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-4.7-flash" in model_id_lower:
elif "glm-5" in model_id_lower or "glm-4.7" in model_id_lower:
# For GLM-5 and GLM-4.7
# 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

View File

@@ -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, 1)
await to_thread.run_sync(self.runner_process.join, 5)
rc = self.runner_process.exitcode
logger.info(f"RunnerSupervisor exited with exit code {rc}")
if rc == 0:

18
uv.lock generated
View File

@@ -378,7 +378,7 @@ dependencies = [
{ 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.dev20260217+50487b41", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#50487b4141f3c951122655db3b83df5146c1fbeb" }, marker = "sys_platform == 'darwin'" },
{ 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-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'" },
@@ -418,7 +418,7 @@ requires-dist = [
{ 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", extras = ["cpu"], marker = "sys_platform == 'linux'", specifier = "==0.30.6" },
{ name = "mlx-lm", specifier = "==0.30.6" },
{ name = "mlx-lm", specifier = "==0.30.7" },
{ name = "msgspec", specifier = ">=0.19.0" },
{ name = "openai-harmony", specifier = ">=0.0.8" },
{ name = "pillow", specifier = ">=11.0,<12.0" },
@@ -1021,7 +1021,7 @@ dependencies = [
{ 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.dev20260217+50487b41", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#50487b4141f3c951122655db3b83df5146c1fbeb" }, marker = "sys_platform == 'darwin'" },
{ 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 = "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'" },
@@ -1068,8 +1068,8 @@ cuda13 = [
[[package]]
name = "mlx"
version = "0.30.7.dev20260217+50487b41"
source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#50487b4141f3c951122655db3b83df5146c1fbeb" }
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'",
]
@@ -1100,20 +1100,20 @@ wheels = [
[[package]]
name = "mlx-lm"
version = "0.30.6"
version = "0.30.7"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "jinja2", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
{ name = "mlx", version = "0.30.7.dev20260217+50487b41", source = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks#50487b4141f3c951122655db3b83df5146c1fbeb" }, marker = "sys_platform == 'darwin'" },
{ 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 = "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/76/cb/815deddc8699b1f694d7e1f9cbed52934c03a8b49432c8add72932bb2f0b/mlx_lm-0.30.6.tar.gz", hash = "sha256:807e042d7040268f1b19190b7eaefd8b2efbff5590a65460974ad4225b91dda1", size = 271733, upload-time = "2026-02-04T21:27:45.741Z" }
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" }
wheels = [
{ 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" },
{ 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" },
]
[[package]]