Update app.py

app.py

@@ -200,7 +200,7 @@ try:
     logger.info(f"Loading model {model_name}")
     model = AutoModelForCausalLM.from_pretrained(
         model_name,
-        torch_dtype=torch.float16,
+        torch_dtype=torch.float32,  # Changed to float32 to avoid Half/Float mismatch
         low_cpu_mem_usage=True
     )
     # Quantize the model for faster CPU inference