[CI] Bump mypy version 1.19.1 -> 1.20.2 (#44647)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

.github/mergify.yml

@@ -36,18 +36,6 @@ pull_request_rules:
 
         For future commits, `pre-commit` will run automatically on changed files before each commit.
 
-        > [!TIP]
-        > <details>
-        > <summary>Is <code>mypy</code> failing?</summary>
-        > <br/>
-        > <code>mypy</code> is run differently in CI. If the failure is related to this check, please use the following command to run it locally:
-        >
-        > ```bash
-        > # For mypy (substitute "3.10" with the failing version if needed)
-        > pre-commit run --hook-stage manual mypy-3.10
-        > ```
-        > </details>
-
 - name: comment-dco-failure
   description: Comment on PR when DCO check fails
   conditions:

.pre-commit-config.yaml

@@ -148,33 +148,27 @@ repos:
     language: python
     entry: python tools/pre_commit/generate_nightly_torch_test.py
     files: ^requirements/test/cuda\.(in|txt)$
-  - id: mypy-local
-    name: Run mypy locally for lowest supported Python version
-    entry: python tools/pre_commit/mypy.py 0 "3.10"
-    stages: [pre-commit] # Don't run in CI
+  - id: mypy-3.10 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
+    name: Run mypy for Python 3.10
+    entry: python tools/pre_commit/mypy.py "3.10"
     <<: &mypy_common
       language: python
       types_or: [python, pyi]
       require_serial: true
-      additional_dependencies: ["mypy[faster-cache]==1.19.1", regex, types-cachetools, types-setuptools, types-PyYAML, types-requests, types-torch, pydantic]
-  - id: mypy-3.10 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
-    name: Run mypy for Python 3.10
-    entry: python tools/pre_commit/mypy.py 1 "3.10"
-    <<: *mypy_common
-    stages: [manual] # Only run in CI
+      additional_dependencies: ["mypy==1.20.2", regex, types-cachetools, types-setuptools, types-PyYAML, types-requests, types-torch, pydantic]
   - id: mypy-3.11 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
     name: Run mypy for Python 3.11
-    entry: python tools/pre_commit/mypy.py 1 "3.11"
+    entry: python tools/pre_commit/mypy.py "3.11"
     <<: *mypy_common
     stages: [manual] # Only run in CI
   - id: mypy-3.12 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
     name: Run mypy for Python 3.12
-    entry: python tools/pre_commit/mypy.py 1 "3.12"
+    entry: python tools/pre_commit/mypy.py "3.12"
     <<: *mypy_common
     stages: [manual] # Only run in CI
   - id: mypy-3.13 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
     name: Run mypy for Python 3.13
-    entry: python tools/pre_commit/mypy.py 1 "3.13"
+    entry: python tools/pre_commit/mypy.py "3.13"
     <<: *mypy_common
     stages: [manual] # Only run in CI
   - id: shellcheck

AGENTS.md

@@ -98,7 +98,7 @@ pre-commit run --all-files
 pre-commit run ruff-check --all-files
 
 # Run mypy as it is in CI:
-pre-commit run mypy-3.10 --all-files --hook-stage manual
+pre-commit run mypy-3.12 --all-files --hook-stage manual
 ```
 
 The line length limit for Python code is 88 characters. If you are not sure, use pre-commit to check.

docs/contributing/README.md

@@ -101,7 +101,7 @@ vLLM's `pre-commit` hooks will now run automatically every time you commit.
     Some `pre-commit` hooks only run in CI. If you need to, you can run them locally with:
 
     ```bash
-    pre-commit run --hook-stage manual mypy-3.10
+    pre-commit run --hook-stage manual mypy-3.11
     ```
 
 ### Documentation

tests/compile/test_graph_partition.py

@@ -565,6 +565,8 @@ def test_size_used_in_multiple_consumer_subgraphs():
     torch._dynamo.mark_dynamic(x, 0)
     torch._dynamo.mark_dynamic(y, 0)
     torch.compile(model_fn, backend=capturing_backend)(x, y)
+    assert captured_graph is not None, "Graph should be captured by backend"
+    assert captured_inputs is not None, "Example inputs should be captured by backend"
 
     split_gm, split_items = split_graph(captured_graph, ["aten::sigmoid"])
 

tests/models/language/pooling/test_colbert.py

@@ -14,7 +14,7 @@ from vllm.entrypoints.pooling.scoring.utils import compute_maxsim_score
 # -----------------------------------------------------------------------
 # Model definitions: (model_name, colbert_dim, extra vllm_runner kwargs)
 # -----------------------------------------------------------------------
-COLBERT_MODELS = {
+COLBERT_MODELS: dict[str, dict] = {
     "bert": {
         "model": "answerdotai/answerai-colbert-small-v1",
         "colbert_dim": 96,

tools/pre_commit/mypy.py

@@ -8,11 +8,9 @@ on files that have been changed. It groups files into different mypy calls
 based on their directory to avoid import following issues.
 
 Usage:
-    python tools/pre_commit/mypy.py <ci> <python_version> <changed_files...>
+    python tools/pre_commit/mypy.py <python_version> <changed_files...>
 
 Args:
-    ci: "1" if running in CI, "0" otherwise. In CI, follow_imports is set to
-        "silent" for the main group of files.
     python_version: Python version to use (e.g., "3.10") or "local" to use
         the local Python version.
     changed_files: List of changed files to check.
@@ -98,8 +96,8 @@ def mypy(
 
 
 def main():
-    python_version = sys.argv[2]
-    file_groups = group_files(sys.argv[3:])
+    python_version = sys.argv[1]
+    file_groups = group_files(sys.argv[2:])
 
     if python_version == "local":
         python_version = f"{sys.version_info.major}.{sys.version_info.minor}"

vllm/model_executor/layers/fused_moe/experts/trtllm_mxint4_moe.py

@@ -30,6 +30,8 @@ class TrtLlmMxint4ExpertsMonolithic(mk.FusedMoEExpertsMonolithic):
         self,
         moe_config: FusedMoEConfig,
         quant_config: FusedMoEQuantConfig,
+        max_num_tokens: int | None = None,
+        num_dispatchers: int | None = None,
     ):
         super().__init__(moe_config, quant_config)
         self.topk = moe_config.experts_per_token

vllm/model_executor/model_loader/tensorizer.py

@@ -687,7 +687,7 @@ def serialize_vllm_model(
         serializer = TensorSerializer(
             stream,
             encryption=encryption_params,
-            **tensorizer_config.serialization_kwargs,
+            **(tensorizer_config.serialization_kwargs or {}),
         )
         serializer.write_module(model)
         serializer.close()