Fix tool_parser_cls type annotation from Callable to type[ToolParser] (#37957)

Signed-off-by: sfeng33 <4florafeng@gmail.com>

vllm/entrypoints/openai/engine/serving.py

@@ -4,7 +4,7 @@ import asyncio
 import contextlib
 import json
 import time
-from collections.abc import AsyncGenerator, Callable, Mapping
+from collections.abc import AsyncGenerator, Mapping
 from dataclasses import dataclass, field
 from http import HTTPStatus
 from typing import Any, ClassVar, Generic, Protocol, TypeAlias, TypeVar
@@ -882,7 +882,7 @@ class OpenAIServing:
         request: ResponsesRequest | ChatCompletionRequest,
         tokenizer: TokenizerLike | None,
         enable_auto_tools: bool,
-        tool_parser_cls: Callable[[TokenizerLike], ToolParser] | None,
+        tool_parser_cls: type[ToolParser] | None,
         content: str | None = None,
     ) -> tuple[list[FunctionCall] | None, str | None]:
         function_calls = list[FunctionCall]()

vllm/entrypoints/openai/parser/responses_parser.py

@@ -39,7 +39,7 @@ class ResponsesParser:
         reasoning_parser_cls: Callable[[TokenizerLike], ReasoningParser],
         response_messages: list[ResponseInputOutputItem],
         request: ResponsesRequest,
-        tool_parser_cls: Callable[[TokenizerLike], ToolParser] | None,
+        tool_parser_cls: type[ToolParser] | None,
     ):
         self.response_messages: list[ResponseInputOutputItem] = (
             # TODO: initial messages may not be properly typed

vllm/entrypoints/openai/responses/context.py

@@ -276,7 +276,7 @@ class ParsableContext(ConversationContext):
         reasoning_parser_cls: Callable[[TokenizerLike], ReasoningParser] | None,
         request: ResponsesRequest,
         available_tools: list[str] | None,
-        tool_parser_cls: Callable[[TokenizerLike], ToolParser] | None,
+        tool_parser_cls: type[ToolParser] | None,
         chat_template: str | None,
         chat_template_content_format: ChatTemplateContentFormatOption,
     ):

vllm/entrypoints/openai/responses/serving.py

@@ -602,7 +602,7 @@ class OpenAIServingResponses(OpenAIServing):
         request: ResponsesRequest,
         messages: list[ResponseInputOutputItem],
         tool_dicts: list[dict[str, Any]] | None,
-        tool_parser: Callable[[TokenizerLike], ToolParser] | None,
+        tool_parser: type[ToolParser] | None,
         chat_template: str | None,
         chat_template_content_format: ChatTemplateContentFormatOption,
     ):

vllm/entrypoints/pooling/base/io_processor.py

@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
-from collections.abc import Callable, Sequence
+from collections.abc import Sequence
 from typing import Any, Final
 
 from vllm import PoolingRequestOutput, PromptType
@@ -21,7 +21,6 @@ from vllm.entrypoints.pooling.typing import (
 from vllm.inputs.data import ProcessorInputs, SingletonPrompt
 from vllm.renderers import BaseRenderer, merge_kwargs
 from vllm.renderers.inputs.preprocess import parse_model_prompt, prompt_to_seq
-from vllm.tokenizers import TokenizerLike
 from vllm.tool_parsers import ToolParser
 from vllm.utils.mistral import is_mistral_tokenizer
 
@@ -167,7 +166,7 @@ class PoolingIOProcessor:
         default_template_content_format: ChatTemplateContentFormatOption,
         default_template_kwargs: dict[str, Any] | None,
         tool_dicts: list[dict[str, Any]] | None = None,
-        tool_parser: Callable[[TokenizerLike], ToolParser] | None = None,
+        tool_parser: type[ToolParser] | None = None,
     ) -> tuple[list[ConversationMessage], list[ProcessorInputs]]:
         renderer = self.renderer
 

vllm/entrypoints/serve/render/serving.py

@@ -1,6 +1,6 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-from collections.abc import Callable, Sequence
+from collections.abc import Sequence
 from http import HTTPStatus
 from typing import Any
 
@@ -45,7 +45,6 @@ from vllm.renderers.inputs.preprocess import (
     parse_model_prompt,
     prompt_to_seq,
 )
-from vllm.tokenizers import TokenizerLike
 from vllm.tool_parsers import ToolParser
 from vllm.utils import random_uuid
 from vllm.utils.mistral import is_mistral_tokenizer
@@ -84,12 +83,10 @@ class OpenAIServingRender:
         self.trust_request_chat_template = trust_request_chat_template
         self.enable_auto_tools = enable_auto_tools
         self.exclude_tools_when_tool_choice_none = exclude_tools_when_tool_choice_none
-        self.tool_parser: Callable[[TokenizerLike], ToolParser] | None = (
-            ParserManager.get_tool_parser(
-                tool_parser_name=tool_parser,
-                enable_auto_tools=enable_auto_tools,
-                model_name=model_config.model,
-            )
+        self.tool_parser: type[ToolParser] | None = ParserManager.get_tool_parser(
+            tool_parser_name=tool_parser,
+            enable_auto_tools=enable_auto_tools,
+            model_name=model_config.model,
         )
         self.default_chat_template_kwargs: dict[str, Any] = (
             default_chat_template_kwargs or {}
@@ -499,7 +496,7 @@ class OpenAIServingRender:
         default_template_content_format: ChatTemplateContentFormatOption,
         default_template_kwargs: dict[str, Any] | None,
         tool_dicts: list[dict[str, Any]] | None = None,
-        tool_parser: Callable[[TokenizerLike], ToolParser] | None = None,
+        tool_parser: type[ToolParser] | None = None,
     ) -> tuple[list[ConversationMessage], list[ProcessorInputs]]:
         """Copied from OpenAIServing._preprocess_chat."""
         renderer = self.renderer