[None][fix] MiniMax-M3.1 DSpark AL parity: non-causal draft blocks + opt-in golden sampling - #21

[None][fix] MiniMax-M3.1 DSpark AL parity: non-causal draft blocks + opt-in golden sampling#21

zheyuf merged 2 commits into

brb-nv:user/brb/m3.1-flash-gb300-initialbrb-nv/TensorRT-LLM:user/brb/m3.1-flash-gb300-initialfrom

zheyuf:zheyuf/brb-dspark-al-parityzheyuf/TensorRT-LLM:zheyuf/brb-dspark-al-parityCopy head branch name to clipboard