Update app.py

app.py

@@ -48,7 +48,7 @@ def _chat_stream(model, tokenizer, query, history, temperature, top_p, max_outpu
     }
 
     try:
-        API_URL = "http://8.152.0.142:8080/v1/chat/completions"
+        API_URL = "http://8.152.0.142:10021/v1/chat/completions"
         response = requests.post(API_URL, headers=headers, data=json.dumps(payload), timeout=60, stream=True)
         response.raise_for_status()
         for chunk in get_streaming_response(response):