Skip to content

Pull requests: alibaba/rtp-llm

New pull request New
55 Open 642 Closed
55 Open 642 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

Develop/bailian
#813 opened Mar 22, 2026 by xinfei-shi Loading…
1
fix - fix qwen35 moe reasoning parser
#812 opened Mar 20, 2026 by alibaba-miji Loading…
2
add emb_dim in modelConfig
#811 opened Mar 20, 2026 by yinjuncheng Loading…
2
fix: 增加checkout步骤超时并优化参数配置
#810 opened Mar 20, 2026 by guoj14 Loading…
1
fix: remove redundent prefill when n > 1 and no beam search
#809 opened Mar 20, 2026 by zhangjianning-zjn Loading…
3
fix: fix module_mha_batch_prefill
#808 opened Mar 20, 2026 by liaocz Loading…
3
feat: use rtp-kernel fp8 group gemm
#807 opened Mar 20, 2026 by moui0 Loading…
4
fix: Skip cudagraph capture at prefillWarmup stage
#805 opened Mar 20, 2026 by bppps Loading…
3
feat: duplicated kv cache.
#804 opened Mar 19, 2026 by ZhangZhiPku Loading…
2
[WIP] feat: support xgrammer
#803 opened Mar 19, 2026 by wanglining97 Loading…
1
Feat/remove cpp device
#802 opened Mar 19, 2026 by JackTan25 Loading…
11
implement kvcache refactor
#801 opened Mar 19, 2026 by ZhihanYan Loading…
20
fix: qwen35 dense model weight loading
#799 opened Mar 18, 2026 by bppps Loading…
2
fix - use triton kernel to prepare cuda graph
#798 opened Mar 18, 2026 by zerozw Loading…
2
perf: only broadcast combo_tokens after draft model forward
#797 opened Mar 18, 2026 by Vinkle-hzt Loading…
6
Feature/p2p connector 3
#795 opened Mar 17, 2026 by zhangchicc Loading…
5
SwapAB + Async Load Cache
#794 opened Mar 17, 2026 by alibaba-miji Loading…
4 tasks done
11
Feat/support sparse cp reuse cache
#792 opened Mar 17, 2026 by MMadhatter Loading…
1
Develop/logprobs opt
#791 opened Mar 17, 2026 by zongyuanwu Loading…
15
feat: support flashinfer qkrmsnorm
#789 opened Mar 17, 2026 by Bruce-Lee-LY Loading…
3
feat: support reuse kvcache within queries with epoch-based cache iso…
#786 opened Mar 17, 2026 by siluzhou Loading…
4
Feat/support sparse cp
#780 opened Mar 14, 2026 by Nancheng-11 Loading…
8
feat: support py flashinfer decode cuda graph and remove c++ flashinfer op
#778 opened Mar 13, 2026 by JackTan25 Loading…
Feat/support gqa cp reuse cache
#770 opened Mar 11, 2026 by MMadhatter Loading…
5
feat:refactor graph to support rocm backend
#761 opened Mar 10, 2026 by muse-coder Loading…
20
ProTip! Follow long discussions with comments:>50.