Skip to content

[TRTLLM-16460][perf] Qwen3.8-Flash-Next NVFP4 modeling_v2 target: GB300 Pareto optimizations for ISL 8K / OSL 1K - #20937

Open
Wanli-Jiang wants to merge 6 commits into
NVIDIA:mainfrom
Wanli-Jiang:user/williamj/qwen38fn-gb300-pareto
Open

Wanli-Jiang wants to merge 6 commits into
NVIDIA:mainfrom
Wanli-Jiang:user/williamj/qwen38fn-gb300-pareto

Commits

  1. Commits on Oct 30, 2026