Format & Improve docs

README.md

@@ -238,6 +238,7 @@ CUDA_VISIBLE_DEVICES=1 python3 -m fastchat.serve.model_worker --model-path lmsys
 ```bash
 python3 -m fastchat.serve.gradio_web_server_multi
 ```
+- The default model worker based on huggingface/transformers has great compatibility but can be slow. If you want high-throughput serving, you can try [vLLM integration](docs/vllm_integration.md).
 
 ## API
 ### OpenAI-Compatible RESTful APIs & SDK

fastchat/serve/controller.py

@@ -337,12 +337,12 @@ def create_controller():
     args, controller = create_controller()
     if args.ssl:
         uvicorn.run(
-            app, 
-            host=args.host, 
-            port=args.port, 
-            log_level="info", 
-            ssl_keyfile=os.environ["SSL_KEYFILE"], 
-            ssl_certfile=os.environ["SSL_CERTFILE"]
+            app,
+            host=args.host,
+            port=args.port,
+            log_level="info",
+            ssl_keyfile=os.environ["SSL_KEYFILE"],
+            ssl_certfile=os.environ["SSL_CERTFILE"],
         )
     else:
         uvicorn.run(app, host=args.host, port=args.port, log_level="info")