Wrap get_model inside the vllm config context#1950

Lumosis merged 1 commit intomainvllm-project/tpu-inference:mainfrom

lihao/oncallvllm-project/tpu-inference:lihao/oncallCopy head branch name to clipboard