Skip to content

Pull requests: alibaba/rtp-llm

New pull request New
89 Open 453 Closed
89 Open 453 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

test: trigger rebuild for AMD CI testing
#658 opened Jan 30, 2026 by Zoe923 Loading…
fix - w13 fused
#657 opened Jan 30, 2026 by jianglan89 Loading…
feat: Support DeepEP dual mode for dynamic Normal/LowLatency switching in PDFusion
#655 opened Jan 30, 2026 by Zoe923 Loading…
16
add comment optimize
#654 opened Jan 29, 2026 by guoj14 Loading…
fix: fix deepep base setting
#648 opened Jan 28, 2026 by Vinkle-hzt Loading…
4
[Feature] Support Per-Expert-Overlap(PEO) capability
#643 opened Jan 27, 2026 by yykzjh Loading…
feat: qwen2.5-vl vit amd merge gate+up, silu_and_mul
#642 opened Jan 27, 2026 by CrimsonDump Loading…
3
support per_group_fp4 weight load and fp4-moe ops
#639 opened Jan 27, 2026 by junna2016 Loading…
18
Revert "chore: rm useless device sync"
#638 opened Jan 27, 2026 by Vinkle-hzt Loading…
3
fix: refactor symm memory to avoid capture oom
#634 opened Jan 26, 2026 by JackTan25 Loading…
8
feat: support headwise attention 260126
#631 opened Jan 26, 2026 by qqbbiu Loading…
1
Feature/sm100 attention
#630 opened Jan 26, 2026 by zerozw Loading…
17
feat: add HybridKVCacheAllocator
#629 opened Jan 25, 2026 by SJTUGavinLiu Loading…
21
Feature/p2p connector
#627 opened Jan 23, 2026 by zhangchicc Loading…
feat: return raw output in streaming mode
#626 opened Jan 23, 2026 by soaringk Loading…
9
fix: fix truncate token type err
#625 opened Jan 23, 2026 by wanglining97 Loading…
feat: use rtp-kernel fused rope kvcache ops for cuda12.9
#623 opened Jan 22, 2026 by moui0 Loading…
47
fix: fix health check error
#622 opened Jan 22, 2026 by ABNER-1 Loading…
feat - c++ flashinfer add headdim=256 && fix xqa log warnning
#620 opened Jan 22, 2026 by alibaba-miji Loading…
fix: fix master role type err
#619 opened Jan 22, 2026 by wanglining97 Loading…
2
chore: skip check for torch profiler get info
#614 opened Jan 21, 2026 by Vinkle-hzt Loading…
2
add model support for deepseek_vl2
#612 opened Jan 21, 2026 by wuchen-shendaxia Loading…
2
feat: support headwise attention
#609 opened Jan 21, 2026 by Echo-2334 Loading…
optimize AMD layernorm
#606 opened Jan 20, 2026 by yanglf1121 Loading…
optimize kv_block_array
#604 opened Jan 20, 2026 by Xu-Sheng-lin Loading…
ProTip! Adding no:label will show everything without a label.