Skip to content

perf(qwen3_5): prefill text prompts in multi-token chunks - #2191

Open
chaofengw-nv wants to merge 3 commits into
NVIDIA:mainfrom
chaofengw-nv:perf/qwen3-5-chunked-prefill
Open

perf(qwen3_5): prefill text prompts in multi-token chunks#2191
chaofengw-nv wants to merge 3 commits into
NVIDIA:mainfrom
chaofengw-nv:perf/qwen3-5-chunked-prefill

Commits

  1. Commits on Nov 10, 2026

  2. Commits on Nov 11, 2026