[Model] Optional FP8 lm_head compression for Llama and Mistral#35696

lucaspirola wants to merge 4 commits intovllm-project:mainvllm-project/vllm:mainfrom

lucaspirola:feat/fp8-lm-head-compresslucaspirola/vllm:feat/fp8-lm-head-compressCopy head branch name to clipboard