Update app.py

app.py

@@ -35,7 +35,7 @@ def generate_response(prompt):
         # Stream tokens for faster perceived response
         for output in model.generate(
             **inputs,
-            max_new_tokens=200,  # Reduced for faster inference
+            max_new_tokens=400,  # Reduced for faster inference
             do_sample=False,     # Greedy decoding for speed
             eos_token_id=tokenizer.eos_token_id,
             pad_token_id=tokenizer.pad_token_id