fix typo quantization

add commandline arg

Author: asaiacai

fastchat/serve/vllm_worker.py

@@ -192,6 +192,7 @@ async def api_model_details(request: Request):
         "--controller-address", type=str, default="http://localhost:21001"
     )
     parser.add_argument("--model-path", type=str, default="lmsys/vicuna-7b-v1.3")
+    parser.add_argument("--quantization", type=str)
     parser.add_argument(
         "--model-names",
         type=lambda s: s.split(","),
@@ -210,7 +211,7 @@ async def api_model_details(request: Request):
         args.model = args.model_path
     if args.num_gpus > 1:
         args.tensor_parallel_size = args.num_gpus
-    if args.quantizaiton:
+    if args.quantization:
         args.quantization = args.quantization
 
     engine_args = AsyncEngineArgs.from_cli_args(args)