[None][Fix] Enable INT8 weight-only (W8A16) MoE for non-gated activations#15550

Dorijan10 wants to merge 1 commit into

NVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom

Dorijan10:feat/int8-woq-moe-nongatedDorijan10/TensorRT-LLM-Optimisation:feat/int8-woq-moe-nongatedCopy head branch name to clipboard