Update app.py

app.py

@@ -33,8 +33,8 @@ def load_model():
 
 pipe = load_model()
 
-@spaces.GPU(duration=110)
-def generate_response(prompt, max_length=1024):
+@spaces.GPU(duration=20)
+def generate_response(prompt, max_length=124):
     # Create messages with system prompt
     messages = [
         {"role": "system", "content": "You are a helpful AI assistant. You always think step by step."},
@@ -42,7 +42,8 @@ def generate_response(prompt, max_length=1024):
     ]
     
     outputs = pipe(messages, max_new_tokens=max_length)
-    
+
+    print("Raw output:", outputs)
     # Extract just the assistant's response
     try:
         # Get the message list from the output