Update app.py

app.py

@@ -5,7 +5,7 @@ client = InferenceClient(model="https://zgg3nzdpswxy4a-80.proxy.runpod.net")
 
 def inference(message, history):
     partial_message = ""
-    for token in client.text_generation(prompt=message, max_new_tokens=512, stream=True, best_of=1, temperature=0.1, watermark=True, 
+    for token in client.text_generation(prompt=message, max_new_tokens=128, stream=True, best_of=1, temperature=0.1, watermark=True, 
                                         top_p=0.95, do_sample=True, repetition_penalty=1.03):
         if token.startswith("<s>"):
             return partial_message