Update llava_llama3/serve/cli.py

llava_llama3/serve/cli.py

@@ -70,7 +70,7 @@ def chat_llava(args, image_file, text, tokenizer, model, image_processor, contex
             do_sample=True if args.temperature > 0 else False,
             temperature=args.temperature,
             max_new_tokens=args.max_new_tokens,
-            # streamer=streamer,
+            streamer=streamer,
             use_cache=True)
 
     outputs = tokenizer.decode(output_ids[0]).strip()