Reduce max_tokens to 100 for faster response to prevent Vapi timeout

app.py

@@ -143,7 +143,7 @@ def chat_completions():
         print(f"Generating response...")
         outputs = model.generate(
             **inputs,
-            max_new_tokens=150,  # Reduced from 256 for faster response
+            max_new_tokens=100,  # Further reduced for faster response (was 150)
             temperature=0.7,
             do_sample=True,
             top_p=0.9,