replace model ID string matching with architecture-based checks in tokenizer loading

Use config.json model_type field instead of fragile substring matching on model IDs for architecture detection in tokenizer loading. Falls back to string matching when config.json is unavailable. Closes #1371 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-18 14:55:13 -05:00 · 2026-02-17 10:19:09 -08:00
10 changed files with 107 additions and 91 deletions
--- a/README.md
+++ b/README.md
@@ -72,23 +72,16 @@ There are two ways to run exo:

 ### Run from Source (macOS)

-If you have [Nix](https://nixos.org/) installed, you can skip most of the steps below and run exo directly (after accepting the Cachix cache):
-
-```bash
-nix run .#exo
-```
-
 **Prerequisites:**
- [Xcode](https://developer.apple.com/xcode/) (provides the Metal ToolChain required for MLX compilation)
 - [brew](https://github.com/Homebrew/brew) (for simple package management on macOS)
-
+  
  ```bash
  /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
  ```
 - [uv](https://github.com/astral-sh/uv) (for Python dependency management)
 - [macmon](https://github.com/vladkens/macmon) (for hardware monitoring on Apple Silicon)
 - [node](https://github.com/nodejs/node) (for building the dashboard)
-
+  
  ```bash
  brew install uv macmon node
  ```
--- a/flake.nix
+++ b/flake.nix
@@ -115,7 +115,7 @@
          packages = lib.optionalAttrs pkgs.stdenv.hostPlatform.isDarwin (
            let
              uvLock = builtins.fromTOML (builtins.readFile ./uv.lock);
-              mlxPackage = builtins.head (builtins.filter (p: p.name == "mlx" && p.source ? git) uvLock.package);
+              mlxPackage = builtins.head (builtins.filter (p: p.name == "mlx") uvLock.package);
              uvLockMlxVersion = mlxPackage.version;
            in
            {
--- a/nix/mlx.nix
+++ b/nix/mlx.nix
@@ -41,16 +41,16 @@ let

  mlx = stdenv.mkDerivation rec {
    pname = "mlx";
-    version = let v = "0.30.7.dev20260217+50487b41"; in
+    version = let v = "0.30.6"; in
      assert v == uvLockMlxVersion || throw "MLX version mismatch: nix/mlx.nix has ${v} but uv.lock has ${uvLockMlxVersion}. Update both the version and hash in nix/mlx.nix.";
      v;
    pyproject = true;

    src = fetchFromGitHub {
-      owner = "rltakashige";
-      repo = "mlx-jaccl-fix-small-recv";
-      rev = "50487b4141f3c951122655db3b83df5146c1fbeb";
-      hash = "sha256-IL4a9vMX5nocgJU1WG4zE8hArHkHJtnh4sdYh3od5zU=";
+      owner = "ml-explore";
+      repo = "mlx";
+      tag = "v${version}";
+      hash = "sha256-avD5EGhwgmPdXLAyQSqTO6AXk/W3ziH+f6AetjK3Sdo=";
    };

    patches = [
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -17,7 +17,7 @@ dependencies = [
    "loguru>=0.7.3",
    "exo_pyo3_bindings", # rust bindings
    "anyio==4.11.0",
-    "mlx; sys_platform == 'darwin'",
+    "mlx==0.30.6; sys_platform == 'darwin'",
    "mlx[cpu]==0.30.6; sys_platform == 'linux'",
    "mlx-lm==0.30.6",
    "tiktoken>=0.12.0", # required for kimi k2 tokenizer
@@ -64,7 +64,6 @@ members = [

 [tool.uv.sources]
 exo_pyo3_bindings = { workspace = true }
-mlx = { git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git", branch = "address-rdma-gpu-locks", marker = "sys_platform == 'darwin'" }
 #mlx-lm = { git = "https://github.com/davidmcc73/mlx-lm", branch = "stable" }
 # Uncomment to use local mlx/mlx-lm development versions:
 # mlx = { path = "/Users/Shared/mlx", editable=true }
--- a/python/parts.nix
+++ b/python/parts.nix
@@ -58,21 +58,6 @@
        lib.optionalAttrs pkgs.stdenv.hostPlatform.isLinux (
          (lib.mapAttrs (_: ignoreMissing) nvidiaPackages) // {
            mlx = ignoreMissing prev.mlx;
-            mlx-cuda-13 = prev.mlx-cuda-13.overrideAttrs (old: {
-              buildInputs = (old.buildInputs or [ ]) ++ [
-                final.nvidia-cublas
-                final.nvidia-cuda-nvrtc
-                final.nvidia-cudnn-cu13
-                final.nvidia-nccl-cu13
-              ];
-              preFixup = ''
-                addAutoPatchelfSearchPath ${final.nvidia-cublas}
-                addAutoPatchelfSearchPath ${final.nvidia-cuda-nvrtc}
-                addAutoPatchelfSearchPath ${final.nvidia-cudnn-cu13}
-                addAutoPatchelfSearchPath ${final.nvidia-nccl-cu13}
-              '';
-              autoPatchelfIgnoreMissingDeps = [ "libcuda.so.1" ];
-            });
            torch = ignoreMissing prev.torch;
            triton = ignoreMissing prev.triton;
          }
@@ -89,25 +74,14 @@
          linuxOverlay
        ]
      );
-      # mlx-cpu and mlx-cuda-13 both ship mlx/ site-packages files; keep first.
-      # mlx-cpu/mlx-cuda-13 and nvidia-cudnn-cu12/cu13 ship overlapping files.
-      venvCollisionPaths = lib.optionals pkgs.stdenv.hostPlatform.isLinux [
-        "lib/python3.13/site-packages/mlx*"
-        "lib/python3.13/site-packages/nvidia*"
-      ];
-
-      exoVenv = (pythonSet.mkVirtualEnv "exo-env" workspace.deps.default).overrideAttrs {
-        venvIgnoreCollisions = venvCollisionPaths;
-      };
+      exoVenv = pythonSet.mkVirtualEnv "exo-env" workspace.deps.default;

      # Virtual environment with dev dependencies for testing
-      testVenv = (pythonSet.mkVirtualEnv "exo-test-env" (
+      testVenv = pythonSet.mkVirtualEnv "exo-test-env" (
        workspace.deps.default // {
          exo = [ "dev" ]; # Include pytest, pytest-asyncio, pytest-env
        }
-      )).overrideAttrs {
-        venvIgnoreCollisions = venvCollisionPaths;
-      };
+      );

      mkPythonScript = name: path: pkgs.writeShellApplication {
        inherit name;
--- a/src/exo/master/api.py
+++ b/src/exo/master/api.py
@@ -143,12 +143,7 @@ from exo.shared.types.openai_responses import (
    ResponsesResponse,
 )
 from exo.shared.types.state import State
-from exo.shared.types.worker.instances import (
-    Instance,
-    InstanceId,
-    InstanceMeta,
-    MlxJacclInstance,
-)
+from exo.shared.types.worker.instances import Instance, InstanceId, InstanceMeta
 from exo.shared.types.worker.shards import Sharding
 from exo.utils.banner import print_startup_banner
 from exo.utils.channels import Receiver, Sender, channel
@@ -472,14 +467,6 @@ class API:
            shard_assignments = instance.shard_assignments
            placement_node_ids = list(shard_assignments.node_to_runner.keys())

-            # Derive instance_meta from the actual instance type, since
-            # place_instance() may override it (e.g., single-node → MlxRing)
-            actual_instance_meta = (
-                InstanceMeta.MlxJaccl
-                if isinstance(instance, MlxJacclInstance)
-                else InstanceMeta.MlxRing
-            )
-
            memory_delta_by_node: dict[str, int] = {}
            if placement_node_ids:
                total_bytes = model_card.storage_size.in_bytes
@@ -492,14 +479,14 @@ class API:
            if (
                model_card.model_id,
                sharding,
-                actual_instance_meta,
+                instance_meta,
                len(placement_node_ids),
            ) not in seen:
                previews.append(
                    PlacementPreview(
                        model_id=model_card.model_id,
                        sharding=sharding,
-                        instance_meta=actual_instance_meta,
+                        instance_meta=instance_meta,
                        instance=instance,
                        memory_delta_by_node=memory_delta_by_node or None,
                        error=None,
@@ -509,7 +496,7 @@ class API:
                (
                    model_card.model_id,
                    sharding,
-                    actual_instance_meta,
+                    instance_meta,
                    len(placement_node_ids),
                )
            )
--- a/src/exo/shared/models/model_cards.py
+++ b/src/exo/shared/models/model_cards.py
@@ -165,6 +165,7 @@ def is_custom_card(model_id: ModelId) -> bool:
 class ConfigData(BaseModel):
    model_config = {"extra": "ignore"}  # Allow unknown fields

+    model_type: str | None = None
    architectures: list[str] | None = None
    hidden_size: Annotated[int, Field(ge=0)] | None = None
    layer_count: int = Field(
@@ -200,6 +201,7 @@ class ConfigData(BaseModel):
            return data

        for field in [
+            "model_type",
            "architectures",
            "hidden_size",
            "num_hidden_layers",
--- a/src/exo/worker/engines/mlx/utils_mlx.py
+++ b/src/exo/worker/engines/mlx/utils_mlx.py
@@ -269,19 +269,52 @@ def get_tokenizer(model_path: Path, shard_metadata: ShardMetadata) -> TokenizerW
    return load_tokenizer_for_model_id(shard_metadata.model_card.model_id, model_path)


-def get_eos_token_ids_for_model(model_id: ModelId) -> list[int] | None:
-    """
-    Get the EOS token IDs for a model based on its ID.
+def _read_model_type_from_config(model_path: Path) -> str | None:
+    """Read the model_type field from config.json at the given model path.

-    Some models require explicit EOS token configuration that isn't in their
-    tokenizer config. This function returns the known EOS token IDs for such models.
+    Returns None if config.json doesn't exist or doesn't contain model_type.
+    """
+    config_path = model_path / "config.json"
+    if not config_path.exists():
+        return None
+    try:
+        with open(config_path) as f:
+            config: dict[str, Any] = json.load(f)  # pyright: ignore[reportAny]
+        model_type: Any = config.get("model_type")
+        if model_type is None:
+            text_config: Any = config.get("text_config")
+            if isinstance(text_config, dict):
+                model_type = text_config.get("model_type")  # pyright: ignore[reportUnknownMemberType,reportUnknownVariableType]
+        return model_type if isinstance(model_type, str) else None
+    except (json.JSONDecodeError, OSError):
+        return None
+
+
+def get_eos_token_ids_for_model(
+    model_id: ModelId, model_type: str | None = None
+) -> list[int] | None:
+    """Get the EOS token IDs for a model based on its architecture type.
+
+    Uses model_type from config.json when available, falls back to model_id
+    string matching for backward compatibility.

    Args:
        model_id: The HuggingFace model ID
+        model_type: The model_type field from config.json (e.g., "kimi", "glm4")

    Returns:
        List of EOS token IDs, or None if the model uses standard tokenizer config
    """
+    if model_type is not None:
+        if model_type == "kimi":
+            return [163586]
+        elif model_type == "glm4_moe_lite":
+            # 154820: <|endoftext|>, 154827: <|user|>, 154829: <|observation|>
+            return [154820, 154827, 154829]
+        elif model_type.startswith("glm"):
+            return [151336, 151329, 151338]
+
+    # Fallback: string matching on model_id
    model_id_lower = model_id.lower()
    if "kimi-k2" in model_id_lower:
        return [163586]
@@ -296,11 +329,10 @@ def get_eos_token_ids_for_model(model_id: ModelId) -> list[int] | None:
 def load_tokenizer_for_model_id(
    model_id: ModelId, model_path: Path
 ) -> TokenizerWrapper:
-    """
-    Load tokenizer for a model given its ID and local path.
+    """Load tokenizer for a model given its ID and local path.

-    This is the core tokenizer loading logic, handling special cases for different
-    model families (Kimi, GLM, etc.) and transformers 5.x compatibility.
+    Uses model_type from config.json for architecture detection when available,
+    falling back to model_id string matching for backward compatibility.

    Args:
        model_id: The HuggingFace model ID (e.g., "moonshotai/Kimi-K2-Instruct")
@@ -309,11 +341,21 @@ def load_tokenizer_for_model_id(
    Returns:
        TokenizerWrapper instance configured for the model
    """
+    model_type = _read_model_type_from_config(model_path)
    model_id_lower = model_id.lower()
-    eos_token_ids = get_eos_token_ids_for_model(model_id)
+    eos_token_ids = get_eos_token_ids_for_model(model_id, model_type=model_type)
+
+    is_kimi = (
+        model_type == "kimi" if model_type is not None else "kimi-k2" in model_id_lower
+    )
+    is_gemma3 = (
+        model_type == "gemma3"
+        if model_type is not None
+        else "gemma-3" in model_id_lower
+    )

    # Kimi uses a custom TikTokenTokenizer that transformers 5.x can't load via AutoTokenizer
-    if "kimi-k2" in model_id_lower:
+    if is_kimi:
        import importlib.util
        import types

@@ -367,7 +409,7 @@ def load_tokenizer_for_model_id(
        eos_token_ids=eos_token_ids,
    )

-    if "gemma-3" in model_id_lower:
+    if is_gemma3:
        gemma_3_eos_id = 1
        gemma_3_end_of_turn_id = 106
        if tokenizer.eos_token_ids is not None:
--- a/src/exo/worker/tests/unittests/test_mlx/test_tokenizers.py
+++ b/src/exo/worker/tests/unittests/test_mlx/test_tokenizers.py
@@ -24,6 +24,7 @@ from exo.worker.engines.mlx.utils_mlx import (

 # Files needed for tokenizer functionality
 TOKENIZER_FILE_PATTERNS = [
+    "config.json",
    "tokenizer.json",
    "tokenizer_config.json",
    "special_tokens_map.json",
@@ -338,6 +339,9 @@ async def test_kimi_tokenizer_specifically():
    # Verify EOS token is set
    assert eos_token_ids == [163586], "Kimi EOS token should be [163586]"

+    # Verify architecture-based detection gives same result
+    assert get_eos_token_ids_for_model(model_id, model_type="kimi") == [163586]
+

 # Test GLM tokenizer since it also has special handling
@pytest.mark.asyncio
@@ -378,3 +382,10 @@ async def test_glm_tokenizer_specifically():
        151329,
        151338,
    ], "GLM EOS tokens should be correct"
+
+    # Verify architecture-based detection gives same result
+    assert get_eos_token_ids_for_model(model_id, model_type="glm4") == [
+        151336,
+        151329,
+        151338,
+    ]
--- a/uv.lock
+++ b/uv.lock
@@ -377,8 +377,8 @@ dependencies = [
    { name = "hypercorn", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "loguru", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "mflux", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.6", source = { registry = "https://pypi.org/simple" }, extra = ["cpu"], marker = "sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.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", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+    { name = "mlx", extra = ["cpu"], marker = "sys_platform == 'linux'" },
    { name = "mlx-lm", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "msgspec", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "openai-harmony", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -416,7 +416,7 @@ requires-dist = [
    { name = "hypercorn", specifier = ">=0.18.0" },
    { name = "loguru", specifier = ">=0.7.3" },
    { name = "mflux", specifier = "==0.15.5" },
-    { name = "mlx", marker = "sys_platform == 'darwin'", git = "https://github.com/rltakashige/mlx-jaccl-fix-small-recv.git?branch=address-rdma-gpu-locks" },
+    { name = "mlx", marker = "sys_platform == 'darwin'", specifier = "==0.30.6" },
    { name = "mlx", extras = ["cpu"], marker = "sys_platform == 'linux'", specifier = "==0.30.6" },
    { name = "mlx-lm", specifier = "==0.30.6" },
    { name = "msgspec", specifier = ">=0.19.0" },
@@ -1020,8 +1020,8 @@ dependencies = [
    { name = "fonttools", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "huggingface-hub", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "matplotlib", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.6", source = { registry = "https://pypi.org/simple" }, extra = ["cuda13"], marker = "sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.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", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+    { name = "mlx", extra = ["cuda13"], marker = "sys_platform == 'linux'" },
    { name = "numpy", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "opencv-python", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
    { name = "piexif", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -1048,12 +1048,18 @@ wheels = [
 name = "mlx"
 version = "0.30.6"
 source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "sys_platform == 'linux'",
+dependencies = [
+    { name = "mlx-metal", marker = "sys_platform == 'darwin'" },
 ]
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/ae/5b/e460e144a34d5529e010056cccf50b538d56ed001473bc6b246018fd58cb/mlx-0.30.6-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:ed86f8bffc174c2f259ca589ea25464c96cf69d1bb457074a2bf2ef53737e54f", size = 573515, upload-time = "2026-02-06T03:45:23.405Z" },
+    { url = "https://files.pythonhosted.org/packages/60/25/69833fefb9a3fef30b56792b1bcd022496c4fea83e45411d289b77ef7546/mlx-0.30.6-cp313-cp313-macosx_15_0_arm64.whl", hash = "sha256:c52294958269e20f300639a17c1900ca8fc737d859ddda737f9811e94bd040e5", size = 573516, upload-time = "2026-02-06T03:45:24.618Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/6a/7e7fbeebc5cb51b6a5eba96b263a6298707bcbdc059f4b0b73e088bc3dea/mlx-0.30.6-cp313-cp313-macosx_26_0_arm64.whl", hash = "sha256:b5b6636f7c49a4d86d8ec82643b972f45a144a7a9f3a967b27b2e6e22cf71e6a", size = 573592, upload-time = "2026-02-06T03:45:25.928Z" },
    { url = "https://files.pythonhosted.org/packages/93/06/280f6f2ba80520a7109730425eda0d966658793aa0d02d8be8d351f75253/mlx-0.30.6-cp313-cp313-manylinux_2_35_aarch64.whl", hash = "sha256:67e6c9e30a9faeacc209917ef5523177cf9b086914b6b5d83ff886e4294b727d", size = 622011, upload-time = "2026-02-06T03:45:28.165Z" },
    { url = "https://files.pythonhosted.org/packages/fe/35/f872afbee9c079cc69924d9e9c46f5663adb7da58cba3511db082dd307c1/mlx-0.30.6-cp313-cp313-manylinux_2_35_x86_64.whl", hash = "sha256:47db8b16fcb6f6c5a47c0bdb24ed377b41237017ac93aa6cb6aa206c9bdf82e4", size = 663650, upload-time = "2026-02-06T03:45:30.315Z" },
+    { url = "https://files.pythonhosted.org/packages/60/23/361dc7a5797634e4d7e9bdd6564c6b28f9b1246672632def2f91bf066b18/mlx-0.30.6-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:78804a89dcff4a838f7c2da72392fe87a523e95122a3c840e53df019122aad45", size = 575028, upload-time = "2026-02-06T03:45:31.549Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/69/1854484d414171586814dfbe8def95f75c4ea2c7341ba13ba8ee675f7c62/mlx-0.30.6-cp314-cp314-macosx_15_0_arm64.whl", hash = "sha256:ec13584ab069665cc7ad34a05494d9291cd623aef6ae96be48875fc87cfc25d6", size = 575026, upload-time = "2026-02-06T03:45:33.072Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/b8/3adbc441924209a7e4c568308b2a0b54bd09aee6a68db5bae85304791e54/mlx-0.30.6-cp314-cp314-macosx_26_0_arm64.whl", hash = "sha256:b2c5e8a090a753ef99a1380a4d059c983083f36198864f6df9faaf1223d083df", size = 575041, upload-time = "2026-02-06T03:45:34.814Z" },
    { url = "https://files.pythonhosted.org/packages/3f/54/9d9e06804fb2088202a2cdf60458e00b221f71420bea285720b60f9e82b5/mlx-0.30.6-cp314-cp314-manylinux_2_35_aarch64.whl", hash = "sha256:9ceddede4af0de31d1f6b3099f70e5469d60cd7c546975dedbdbeab3519cab3f", size = 624002, upload-time = "2026-02-06T03:45:36Z" },
    { url = "https://files.pythonhosted.org/packages/42/92/3140a15a50cb1f9267a6552171e1dfa577861de53e093124bc43707f2a0e/mlx-0.30.6-cp314-cp314-manylinux_2_35_x86_64.whl", hash = "sha256:4a6ffd2d16728cf95f63a1b555d7c2eaeea686a0e6b73228bd265411cb5d77a4", size = 663569, upload-time = "2026-02-06T03:45:37.242Z" },
 ]
@@ -1066,14 +1072,6 @@ cuda13 = [
    { name = "mlx-cuda-13", marker = "sys_platform == 'linux'" },
 ]

-[[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" }
-resolution-markers = [
-    "sys_platform == 'darwin'",
-]
-
 [[package]]
 name = "mlx-cpu"
 version = "0.30.6"
@@ -1104,7 +1102,7 @@ version = "0.30.6"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
    { name = "jinja2", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
-    { name = "mlx", version = "0.30.7.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", 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'" },
@@ -1116,6 +1114,16 @@ 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" },
 ]

+[[package]]
+name = "mlx-metal"
+version = "0.30.6"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f3/85/44406b521f920248fad621334d4dc15e77660a494edf890e7cbee33bf38d/mlx_metal-0.30.6-py3-none-macosx_14_0_arm64.whl", hash = "sha256:ea6d0c973def9a5b4f652cc77036237db3f88c9d0af63701d76b5fddde99b820", size = 38437818, upload-time = "2026-02-06T03:44:56.19Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/cb/10a516995f7d0c154b0d7e633c54b51e96977a86a355105b6474cfcbe0d0/mlx_metal-0.30.6-py3-none-macosx_15_0_arm64.whl", hash = "sha256:0f8cb94634d07e06a372d6ad9a090f38a18bab1ff19a140aede60eacf707bb94", size = 38433701, upload-time = "2026-02-06T03:44:59.678Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/7d/70cb272f7373c334709f210ed8420511fc9d64d05a7a646c0b3b94c29c04/mlx_metal-0.30.6-py3-none-macosx_26_0_arm64.whl", hash = "sha256:d761ae26304f2c4b454eeea7f612a56919d9e5e57dbb1dc0788f8e34aa6f41c2", size = 47718448, upload-time = "2026-02-06T03:45:03.133Z" },
+]
+
 [[package]]
 name = "more-itertools"
 version = "10.8.0"