Skip to content

[Relax][WebGPU] Sequence-sharded paged decode attention for long contexts - #20426

Open
JCNTH wants to merge 2 commits into
apache:mainfrom
JCNTH:webgpu-decode-attention
Open

[Relax][WebGPU] Sequence-sharded paged decode attention for long contexts#20426
JCNTH wants to merge 2 commits into
apache:mainfrom
JCNTH:webgpu-decode-attention

Commits

  1. Commits on Oct 13, 2026

  2. Commits on Nov 15, 2026