[Frontend][Feature] support tool calling for internlm/internlm2_5-7b-chat model (#8405)

vllm/entrypoints/openai/api_server.py

@@ -53,6 +53,7 @@ from vllm.entrypoints.openai.serving_embedding import OpenAIServingEmbedding
 from vllm.entrypoints.openai.serving_engine import BaseModelPath
 from vllm.entrypoints.openai.serving_tokenization import (
     OpenAIServingTokenization)
+from vllm.entrypoints.openai.tool_parsers import ToolParserManager
 from vllm.logger import init_logger
 from vllm.usage.usage_lib import UsageContext
 from vllm.utils import FlexibleArgumentParser, get_open_zmq_ipc_path
@@ -526,6 +527,15 @@ async def run_server(args, **uvicorn_kwargs) -> None:
     logger.info("vLLM API server version %s", VLLM_VERSION)
     logger.info("args: %s", args)
 
+    if args.tool_parser_plugin and len(args.tool_parser_plugin) > 3:
+        ToolParserManager.import_tool_parser(args.tool_parser_plugin)
+
+    valide_tool_parses = ToolParserManager.tool_parsers.keys()
+    if args.enable_auto_tool_choice \
+        and args.tool_call_parser not in valide_tool_parses:
+        raise KeyError(f"invalid tool call parser: {args.tool_call_parser} "
+                       f"(chose from {{ {','.join(valide_tool_parses)} }})")
+
     temp_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
     temp_socket.bind(("", args.port))