[https://nvbugs/6691772][fix] Update trtllm-gen fmha autotuner to enable 96 head MLA - #18599

[https://nvbugs/6691772][fix] Update trtllm-gen fmha autotuner to enable 96 head MLA#18599

pengbowang-nv merged 2 commits into

NVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom

pengbowang-nv:dev-feat-update-trtllmgen-autotuner-for-k3-mlapengbowang-nv/TensorRT-LLM:dev-feat-update-trtllmgen-autotuner-for-k3-mlaCopy head branch name to clipboard