Update app.py

app.py

@@ -25,7 +25,7 @@ def load_model():
 
 model, tokenizer = load_model()
 
-@spaces.GPU(duration=60)  # Added the decorator here
+@spaces.GPU(duration=110)  # Added the decorator here
 def generate_response(prompt, max_length=1024):
     inputs = tokenizer(prompt, return_tensors="pt").to("cuda" if torch.cuda.is_available() else "cpu")
     outputs = model.generate(**inputs, max_length=max_length)