[Quantization] MSE-calibrate every per-expert weight in fused-experts MoE#1421

cjluo-nv merged 2 commits into

mainNVIDIA/Model-Optimizer:mainfrom

chenjiel/recipe-nvfp4-experts-mse-fp8-cast-kv-3NVIDIA/Model-Optimizer:chenjiel/recipe-nvfp4-experts-mse-fp8-cast-kv-3Copy head branch name to clipboard