Update app.py

app.py

@@ -51,8 +51,8 @@ def start_worker(model_path: str, bits=16):
         "llava-hf/llava-1.5-7b-hf",
         "--use-flash-attn",
     ]
-    if bits != 16:
-        worker_command += [f"--load-{bits}bit"]
+    # if bits != 16:
+    #    worker_command += [f"--load-{bits}bit"]
     print(worker_command)
     return subprocess.Popen(worker_command)