Add standalone layer-wise KV-cache AutoQuant with forward KL - #2272

Add standalone layer-wise KV-cache AutoQuant with forward KL#2272

meenchen wants to merge 12 commits into

mainNVIDIA/Model-Optimizer:mainfrom

agent/kv-cache-autoquant-coreNVIDIA/Model-Optimizer:agent/kv-cache-autoquant-coreCopy head branch name to clipboard