"add support for vllm api stop parameter"

Former-commit-id: b9f21fa639b66db09c79404d885661c96bdf9395

src/llmtuner/api/app.py

@@ -141,6 +141,7 @@ def create_app(chat_model: "ChatModel") -> "FastAPI":
             top_p=request.top_p,
             max_new_tokens=request.max_tokens,
             num_return_sequences=request.n,
+            stop=request.stop
         )
 
         prompt_length, response_length = 0, 0
@@ -193,6 +194,7 @@ def create_app(chat_model: "ChatModel") -> "FastAPI":
             temperature=request.temperature,
             top_p=request.top_p,
             max_new_tokens=request.max_tokens,
+            stop=request.stop
         ):
             if len(new_token) == 0:
                 continue

src/llmtuner/api/protocol.py

@@ -1,6 +1,6 @@
 import time
 from enum import Enum, unique
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Union
 
 from pydantic import BaseModel, Field
 from typing_extensions import Literal
@@ -78,6 +78,7 @@ class ChatCompletionRequest(BaseModel):
     n: int = 1
     max_tokens: Optional[int] = None
     stream: bool = False
+    stop: Union[Optional[str], List[str]] = None
 
 
 class ChatCompletionResponseChoice(BaseModel):