Update app.py

app.py

@@ -20,7 +20,7 @@ class StopOnTokens(StoppingCriteria):
                 return True
         return False
 model.to('cuda')
-@spaces.GPU(duration=195)
+@spaces.GPU(duration=300)
 def predict(message, history, temperature, max_tokens, top_p, top_k):
     history_transformer_format = history + [[message, ""]]
     stop = StopOnTokens()