Skip to content

Pull requests: lightseekorg/tokenspeed

New pull request New
16 Open 218 Closed
16 Open 218 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

perf(deepseek-v4): vectorize read_deepseek_v4_indexer_fp8_cache
#238 opened May 24, 2026 by yuanqingz Loading…
5
feat(runtime): support multimodal VLM high priority
#236 opened May 24, 2026 by chenht2022 Contributor Loading…
1 task done
11
Support DP sampling for spec decode
#232 opened May 24, 2026 by yubofredwang Contributor Loading…
9
Use operand format signatures for kernel selection
#230 opened May 23, 2026 by antiagainst Member Loading…
3
[WIP] perf(eagle3): skip dead-position compute in draft catch-up step
#217 opened May 22, 2026 by rjzhb Loading…
1 task done
6
fix(deepseek-v4): close MTP acceptance gap
#207 opened May 21, 2026 by Xiangyi1996 Loading…
17
fix(logits): avoid nan in fused softcap
#183 opened May 19, 2026 by elwhyjay Contributor Loading…
feat(trtllm-MHA): support mixed prefill/decode batches
#176 opened May 18, 2026 by rjzhb Loading…
4 tasks done
7
feat: support post-norm EAGLE + add speculative decoding docs high priority
#174 opened May 17, 2026 by Dogacel Loading…
@yubofredwang
perf(moe): triton biased grouped topk for deepseek-v3 routing
#171 opened May 17, 2026 by roycho96 Contributor Loading…
2
wip: EAGLE post-norm
#170 opened May 17, 2026 by Dogacel • Draft
perf(sampling): opt-in fast verify path for topk=1 chain spec
#133 opened May 13, 2026 by cicirori Collaborator • Draft
4 tasks done
[WIP] feat(lora): LoRA adapter serving
#83 opened May 11, 2026 by qywu Collaborator • Draft
1 of 7 tasks
fix: retraction load back race condition.
#74 opened May 11, 2026 by LorrinWWW Contributor Loading…
perf: chunked-prefill prefix cache update for non-hybrid models
#22 opened May 7, 2026 by LorrinWWW Contributor Loading…
@LorrinWWW
1
fix: wait per-layer on drafter KV pool during cpu cache loadback
#6 opened May 6, 2026 by LorrinWWW Contributor Loading…
@LorrinWWW
ProTip! no:milestone will show everything without a milestone.