A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 10, 09:00 UTC
[FixBug] Improve exception string in `tensorizer.py` (#31680)
Signed-off-by: maang <maang_h@163.com> Signed-off-by: maang-h <55082429+maang-h@users.noreply.github.com> Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
maang committed
Jan 11, 2026 at 21:01 UTC
a34abc49b7553d4af8d60fdeb78b5b5312e4f350
1 file changed
+4
-1
vllm/model_executor/model_loader/tensorizer.py
+4
-1
@@ -764,7 +764,10 @@ def tensorize_lora_adapter(lora_path: str, tensorizer_config: TensorizerConfig):
764
elif tensor_path.endswith(".bin"):
765
tensors = torch.load(tensor_path, weights_only=True)
766
else:
767
- raise ValueError("Unsupported file: %s", tensor_path)
767
+ raise ValueError(
768
+ f"Unsupported adapter model file: {tensor_path}. "
769
+ f"Must be a .safetensors or .bin file."
770
+ )
771
772
with open(config_path) as f:
773
config = json.load(f)