diff --git a/load_models.py b/load_models.py
index b7fa5a37..6822515a 100644
--- a/load_models.py
+++ b/load_models.py
@@ -54,10 +54,11 @@ def load_quantized_model_gguf_ggml(model_id, model_basename, device_type, loggin
             kwargs["n_gpu_layers"] = N_GPU_LAYERS  # set this based on your GPU
 
         return LlamaCpp(**kwargs)
-    except:
+    except Exception as e:
         if "ggml" in model_basename:
             logging.INFO("If you were using GGML model, LLAMA-CPP Dropped Support, Use GGUF Instead")
-        return None
+        logging.info("LlamaCpp conversion error: ")
+        raise e
 
 
 def load_quantized_model_qptq(model_id, model_basename, device_type, logging):