Skip to content

Pull requests: alibaba/rtp-llm

New pull request New
56 Open 587 Closed
56 Open 587 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

fix - fix pywrappedmodel attn object not hold
#759 opened Mar 9, 2026 by zerozw Loading…
Headwise ut
#758 opened Mar 9, 2026 by qqbbiu Loading…
Feature/support qwen35 merge
#755 opened Mar 6, 2026 by alibaba-miji Loading…
feat: remove old sp engine
#753 opened Mar 6, 2026 by JackTan25 Loading…
feat: separate py model from gpt model
#752 opened Mar 6, 2026 by JackTan25 Loading…
feat: refactor reuse cache on rocm python mode
#751 opened Mar 6, 2026 by muse-coder Loading…
feat: support w4a8
#750 opened Mar 6, 2026 by Bruce-Lee-LY Loading…
feat: update the code checkout step and add retry
#749 opened Mar 6, 2026 by guoj14 Loading…
Qwen moe pure tp support
#748 opened Mar 5, 2026 by hxy0118 Loading…
4
feat: support sp prefill cuda graph
#745 opened Mar 5, 2026 by JackTan25 Loading…
Fix typo: 'seperated' -> 'separated' in bench_util.py
#744 opened Mar 5, 2026 by hobostay Loading…
feat: support headwise attention
#742 opened Mar 5, 2026 by Echo-2334 Loading…
fix: add repository checkout retry mechanism
#741 opened Mar 5, 2026 by guoj14 Loading…
Develop/kvcache refactor 3
#739 opened Mar 3, 2026 by xinfei-shi Loading…
Feature/p2p connector merge 1
#735 opened Mar 3, 2026 by zhangchicc Loading…
feat: support swizzle of rocm vit
#734 opened Mar 3, 2026 by missximon Loading…
refactor: use graphbase class to support graph mode on both cuda and rocm backend
#730 opened Mar 2, 2026 by muse-coder Loading…
feat: refactor frontend server
#729 opened Mar 2, 2026 by wanglining97 Loading…
4
fix: refactor kvcache structure and support kernel_block_size
#724 opened Feb 27, 2026 by SJTUGavinLiu Loading…
15
feat: add multi group fmha_impl for hybrid attention
#717 opened Feb 27, 2026 by SJTUGavinLiu Loading…
12
feat: deterministic_gemm for test
#711 opened Feb 26, 2026 by LLLLKKKK Loading…
enable force gather batch & batch isolation
#702 opened Feb 12, 2026 by CHW0218 Loading…
fix: seperate different model flashinfer params
#697 opened Feb 11, 2026 by Vinkle-hzt Loading…
1
feat: MTP-compatible Streaming Parsing detectors
#694 opened Feb 10, 2026 by soaringk Loading…
1
Qwen3 next speculative decoding support
#688 opened Feb 9, 2026 by Vinkle-hzt Loading…
28
ProTip! Adding no:label will show everything without a label.