Skip to content

Pull requests: vllm-project/vllm-ascend

New pull request New
165 Open 1,491 Closed
165 Open 1,491 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

[4/N][Refactor] Extract _prepare_build_info() and _assemble_build_info() from build() in AscendAttentionMetadataBuilder
#2375 opened Aug 14, 2025 by shen-shanshan Loading…
3
[Bugfix][PD]fix non-working disaggregated prefill
#2374 opened Aug 14, 2025 by CaveNightingale • Draft
4
[Qwen-moe] Remove the minor operation arange module:ops module:quantization module:tests
#2373 opened Aug 14, 2025 by s30076806 Loading…
3
[CORE] concurrent partial prefills documentation Improvements or additions to documentation
#2372 opened Aug 14, 2025 by Csrayz Loading…
3
[v0.9.1] Refactor attn metadata build
#2371 opened Aug 14, 2025 by JC-ut0 Loading…
5
refactor allgather/mc2-related fused_experts module:ops module:tests
#2369 opened Aug 14, 2025 by momo609 Loading…
8
[Quickfix] update CachedRequestState as NewRequestData changed
#2367 opened Aug 14, 2025 by MengqingCao Loading…
4
[Feat] support static EPLB for allgather EP module:quantization
#2366 opened Aug 14, 2025 by realliujiaxu Loading…
2
[v0.9.1-dev]dynamic eplb metrics
#2364 opened Aug 14, 2025 by baxingpiaochong Loading…
5
[Refactor][WIP] Refactor mla_v1 by moving all MLA preprocessing ops into mla_v1 attention impl. documentation Improvements or additions to documentation module:core
#2363 opened Aug 14, 2025 by whx-sjtu Loading…
7
[Qwen-moe] use npu_add_rms_norm_quant operator
#2362 opened Aug 14, 2025 by s30076806 Loading…
5
[Refactor] refactor spec decode
#2361 opened Aug 14, 2025 by wangxiyuan Loading…
19
refact attn metadata build
#2360 opened Aug 14, 2025 by weiguihua2 Loading…
11
[0.9.1] fix dbo error info merge-conflicts module:core module:tests
#2358 opened Aug 14, 2025 by ZhaoJiangJiang Loading…
3
Fix the accuracy issues caused by the mrope operator module:ops
#2355 opened Aug 13, 2025 by zhihaofang1017 Loading…
1
[Bugfix]Support Qwen3-MOE on aclgraph mode in sizes capture and add new ut module:core module:tests
#2352 opened Aug 13, 2025 by lilinsiman Loading…
10
Topology-Aware Expert Load Balancing(TA-ELB) optimazation for EPLB module:ops module:quantization
#2351 opened Aug 13, 2025 by Skywalker-EP Loading…
7
[WIP][CI] Add multi node ci
#2345 opened Aug 13, 2025 by Potabk • Draft
6
refact mla forward_prefill&forward_decode merge-conflicts
#2342 opened Aug 12, 2025 by SunnyLee151064 Loading…
4
Delete dumplicate codes and fix a bug module:quantization
#2341 opened Aug 12, 2025 by SunnyLee151064 Loading…
3
[Feat] support return dumped blocks id for connector merge-conflicts
#2340 opened Aug 12, 2025 by flesher0813 Loading…
5
add document for deepseek large EP documentation Improvements or additions to documentation
#2339 opened Aug 12, 2025 by hust17yixuan Loading…
5
Torchair graph812 cov module:core module:ops module:tests
#2337 opened Aug 12, 2025 by NicholasTao Loading…
6
[Misc] Refacotr prepare_reqs merge-conflicts
#2335 opened Aug 12, 2025 by wangxiyuan Loading…
5
[0.9.1] fix error info module:core
#2333 opened Aug 12, 2025 by shikang-hangzhou Loading…
2
ProTip! no:milestone will show everything without a milestone.