feat: add dedicated Tool Call model from Mistral

vllm/entrypoints/openai/protocol.py

@@ -3,6 +3,8 @@
 import time
 from argparse import Namespace
 from typing import Any, Dict, List, Literal, Optional, Union
+from random import choices
+from string import ascii_letters, digits
 
 import torch
 from openai.types.chat import ChatCompletionContentPartParam
@@ -21,6 +23,8 @@
 _MOCK_LONG_INFO = Namespace(min=-9223372036854775808, max=9223372036854775807)
 _LONG_INFO: Union["torch.iinfo", Namespace]
 
+ALPHANUMERIC = ascii_letters + digits
+
 try:
     from sphinx.ext.autodoc.mock import _MockModule
 
@@ -772,6 +776,17 @@ class ToolCall(OpenAIBaseModel):
     function: FunctionCall
 
 
+class MistralToolCall(ToolCall):
+    id: str = Field(
+        default_factory=lambda: MistralToolCall.generate_random_id())
+
+    @staticmethod
+    def generate_random_id():
+        # Mistral Tool Call Ids must be alphanumeric with a maximum length of 9.
+        # https://github.com/mistralai/mistral-common/blob/21ee9f6cee3441e9bb1e6ed2d10173f90bd9b94b/src/mistral_common/protocol/instruct/validator.py#L299
+        return "".join(choices(ALPHANUMERIC, k=9))
+
+
 class DeltaFunctionCall(BaseModel):
     name: Optional[str] = None
     arguments: Optional[str] = None
@@ -790,7 +805,7 @@ class ExtractedToolCallInformation(BaseModel):
     tools_called: bool
 
     # extracted tool calls
-    tool_calls: List[ToolCall]
+    tool_calls: List[ToolCall | MistralToolCall]
 
     # content - per OpenAI spec, content AND tool calls can be returned rarely
     # But some models will do this intentionally
@@ -800,7 +815,7 @@ class ExtractedToolCallInformation(BaseModel):
 class ChatMessage(OpenAIBaseModel):
     role: str
     content: Optional[str] = None
-    tool_calls: List[ToolCall] = Field(default_factory=list)
+    tool_calls: List[ToolCall | MistralToolCall] = Field(default_factory=list)
 
 
 class ChatCompletionLogProb(OpenAIBaseModel):