Skip to content

Pull requests: vllm-project/vllm-ascend

New pull request New
66 Open 720 Closed
66 Open 720 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

Revert break change module:tests
#1198 opened Jun 12, 2025 by wangxiyuan Loading…
[Fix] Revert "Add qwen2.5 vl multimodal feature for vllm-ascend v1 (#736)" module:tests
#1197 opened Jun 12, 2025 by yiz-liu Loading…
2
[EPLB] support deepseek eplb strategy
#1196 opened Jun 12, 2025 by ZhengWG Loading…
[perf] Replace _npu_rotary_embedding with npu_mrope module:ops module:tests
#1195 opened Jun 12, 2025 by David9857 Loading…
Improving performance in graph mode by adjusting the communication me…
#1194 opened Jun 12, 2025 by sharonyunyun Loading…
1
[Feature] Dynamic EPLB Zero-like-overhead Expert Movement merge-conflicts module:core module:ops module:quantization
#1190 opened Jun 12, 2025 by raindaywhu Loading…
1
[Bugfix][Spec Decode] Little fix to spec decode in model_runner_v1.py
#1189 opened Jun 12, 2025 by shen-shanshan • Draft
[v0.9.1] Add docs/source/confg.py print documentation Improvements or additions to documentation
#1187 opened Jun 12, 2025 by Yikun • Draft
add static EPLB unit test module:quantization module:tests
#1186 opened Jun 12, 2025 by songshanhu07 Loading…
Recover scheduler test
#1185 opened Jun 12, 2025 by Yikun Loading…
3
Support Deepseek w4a8 quantization module:quantization
#1182 opened Jun 12, 2025 by pichangping Loading…
3
[CI] Open ut for ascend scheduler only in ci of v1. ready read for review
#1180 opened Jun 12, 2025 by whx-sjtu Loading…
1
[Doctest] add installation doctest documentation Improvements or additions to documentation module:tests
#1179 opened Jun 11, 2025 by Yikun • Draft
1
support fused_experts_allgather_ep module:core module:quantization
#1177 opened Jun 11, 2025 by wgd599 Loading…
4
test Pr/989 merge-conflicts module:core module:tests
#1173 opened Jun 11, 2025 by Yikun • Draft
1
vllm-ascend support chunked prefill documentation Improvements or additions to documentation long-term-test enable long term test for PR module:core module:tests ready-for-test start test by label for PR
#1172 opened Jun 11, 2025 by fems14 Loading…
9
[refactor] Refactoring AscendFusedMoE long-term-test enable long term test for PR module:core module:ops module:quantization ready-for-test start test by label for PR
#1169 opened Jun 11, 2025 by zzzzwwjj Loading…
1
8
Enable code cov for V1 merge-conflicts
#1164 opened Jun 10, 2025 by Yikun • Draft
3
[CI] Add accuracy ci for DP and EP and TP long-term-test enable long term test for PR module:tests ready-for-test start test by label for PR
#1140 opened Jun 9, 2025 by zhangxinyuehfad Loading…
3
[CI] Run deepseek acc ut per pr merge-conflicts module:tests
#1136 opened Jun 9, 2025 by MengqingCao Loading…
5
Waiting for BMM NZ support(Improve TPOP 2ms performance) ready read for review
#1131 opened Jun 9, 2025 by ttanzhiqiang Loading…
1
[V1][Spec-decode] First Stage support of Eagle 1 merge-conflicts module:core
#1128 opened Jun 9, 2025 by umeiko Loading…
1
[docs] Update guidance on how to implement and register new models documentation Improvements or additions to documentation
#1126 opened Jun 9, 2025 by linfeng-yuan Loading…
@shen-shanshan
11
[Feature] Use max_num_seqs tokens with profile_run for decode
#1110 opened Jun 7, 2025 by jianzs • Draft
[perf] optimize apply_penalties & topKtopP for V0&V1 Engine merge-conflicts
#1107 opened Jun 6, 2025 by linfeng-yuan Loading…
2
ProTip! Mix and match filters to narrow down what you’re looking for.