Skip to content

Pull requests: lightseekorg/tokenspeed

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

feat(gemm): add w4a16 linear via flashinfer cutedsl
#1371 opened Sep 2, 2026 by minedec Contributor Draft
perf(topk): optimize persistent radix top-k for long sequences
#1366 opened Sep 2, 2026 by tuanzhangCS Contributor Loading…
feat(ple): support qwen3.8-flash-next PLE cpu offload
#1362 opened Sep 2, 2026 by minedec Contributor Draft
refactor(ple): reorganize kernels and optimize qwen4-exp execution
#1361 opened Sep 2, 2026 by minedec Contributor Loading…
[WIP] Add DeepSeek V4 Gluon sparse attention
#1360 opened Sep 2, 2026 by aditithanekar Loading…
fix(engine): retain aborted requests until terminal ack
#1347 opened Sep 1, 2026 by FlamingoPg Contributor Loading…
test: update k3 agentic bench
#1326 opened Aug 31, 2026 by syuoni Member Draft
perf(kda): use FlashInfer fused decode on Blackwell
#1311 opened Aug 30, 2026 by chenht2022 Contributor Draft
feat: add DFlash2 draft serving support
#1211 opened Aug 23, 2026 by torchspec-bot Collaborator Draft
ProTip! Add no:assignee to see everything that’s not assigned.