vllm
Fix GPTQ model loading in Transformers backend
#25770
Merged

Loading