Add vLLM version info to logs and openai API server (#3161)

vllm/engine/llm_engine.py

@@ -7,6 +7,7 @@
 from typing import (TYPE_CHECKING, Any, Dict, Iterable, List, Optional, Tuple,
                     Union)
 
+import vllm
 from vllm.lora.request import LoRARequest
 from vllm.config import (CacheConfig, DeviceConfig, ModelConfig,
                          ParallelConfig, SchedulerConfig, LoRAConfig)
@@ -85,7 +86,7 @@ def __init__(
         log_stats: bool,
     ) -> None:
         logger.info(
-            "Initializing an LLM engine with config: "
+            f"Initializing an LLM engine (v{vllm.__version__}) with config: "
             f"model={model_config.model!r}, "
             f"tokenizer={model_config.tokenizer!r}, "
             f"tokenizer_mode={model_config.tokenizer_mode}, "

vllm/entrypoints/openai/api_server.py

@@ -15,6 +15,7 @@
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import JSONResponse, StreamingResponse, Response
 
+import vllm
 from vllm.engine.arg_utils import AsyncEngineArgs
 from vllm.engine.async_llm_engine import AsyncLLMEngine
 from vllm.entrypoints.openai.protocol import CompletionRequest, ChatCompletionRequest, ErrorResponse
@@ -168,6 +169,12 @@ async def show_available_models():
     return JSONResponse(content=models.model_dump())
 
 
+@app.get("/version")
+async def show_version():
+    ver = {"version": vllm.__version__}
+    return JSONResponse(content=ver)
+
+
 @app.post("/v1/chat/completions")
 async def create_chat_completion(request: ChatCompletionRequest,
                                  raw_request: Request):
@@ -231,6 +238,7 @@ async def authentication(request: Request, call_next):
                 f"Invalid middleware {middleware}. Must be a function or a class."
             )
 
+    logger.info(f"vLLM API server version {vllm.__version__}")
     logger.info(f"args: {args}")
 
     if args.served_model_name is not None: