Update app.py

app.py

@@ -73,7 +73,7 @@ def respond(
     # Use Qwen Math model for better mathematical reasoning
     client = InferenceClient(
       #  token=hf_token.token if hf_token else None,
-        model="Qwen/Qwen2.5-Math-PRM-72B"  # Specialized math model
+        model="Qwen/Qwen2.5-Math-72B"  # Specialized math model
     )
 
     # Build message history