Skip to content

Pull requests: vllm-project/vllm-ascend

New pull request New
132 Open 1,965 Closed
132 Open 1,965 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

[1/N][Refactor][Qwen3-Next] remove redundant Qwen3NextSparseMoeBlock and Qwen3NextAttention
#3019 opened Sep 18, 2025 by wxsIcey • Draft
1
Refactor ci to reuse base workflow
#3018 opened Sep 18, 2025 by Yikun • Draft
1
enable hierarchical communication for mc2 ops on A2 module:ops module:tests
#3015 opened Sep 18, 2025 by realliujiaxu Loading…
5
[Attention] vllm-ascend support msMonitor tool module:core
#3014 opened Sep 18, 2025 by mfyCn-1204 Loading…
3
[main][quantization] Support deepseek w4a8 per-channel quantization documentation Improvements or additions to documentation module:quantization
#3011 opened Sep 18, 2025 by kunpengW-code Loading…
6
[Bugfix] Fix async copy bug under single expert scenario module:ops
#3005 opened Sep 18, 2025 by sdmyzlp Loading…
6
[Model] Remove deepseek modeling.
#3004 opened Sep 18, 2025 by whx-sjtu Loading…
2
[Refactor] Adjustments to moe_comm_method selection process module:core module:ops module:tests ready read for review ready-for-test start test by label for PR
#3001 opened Sep 18, 2025 by Pr0Wh1teGivee Loading…
6
add a case module:tests
#2995 opened Sep 17, 2025 by jiangyunfan1 Loading…
6
[CustomOp] Register AscendSharedFusedMoE custom op module:core module:ops ready read for review ready-for-test start test by label for PR
#2980 opened Sep 17, 2025 by shen-shanshan Loading…
3
[Bugfix] eagle and eagle3 spec decode failures and enable e2e test module:tests ready read for review ready-for-test start test by label for PR
#2979 opened Sep 17, 2025 by wxsIcey Loading…
7
[Bugfix] ngram spec decode attention error and repeat add sampled token ids error
#2972 opened Sep 17, 2025 by wxsIcey Loading…
5
[Bugfix] Remove VLLM_TEST_DYNAMO_FULLGRAPH_CAPTURE ready read for review ready-for-test start test by label for PR
#2969 opened Sep 16, 2025 by Lucaskabela Loading…
1
1
fix: BSH->TND
#2967 opened Sep 16, 2025 by mikasa955 Loading…
2
fix dp of has_unfinished_dp
#2966 opened Sep 16, 2025 by zhaowei1936 Loading…
3
chunked prefill, access splitfuse op
#2962 opened Sep 16, 2025 by qyqc731 Loading…
5
support cp&sp merge-conflicts module:core module:ops module:tests ready-for-test start test by label for PR
#2961 opened Sep 16, 2025 by LookAround0301 Loading…
8
[feature]GLM4.5 support gatingtopk module:ops
#2958 opened Sep 16, 2025 by 1092626063 Loading…
3
merge common_fused_moe & fused_moe module:core module:ops module:quantization
#2957 opened Sep 16, 2025 by Pr0Wh1teGivee Loading…
3
Add e2e test related to weight updates in RL scenarios. module:tests ready read for review ready-for-test start test by label for PR
#2954 opened Sep 16, 2025 by momo609 Loading…
10
[WIP][Feature] mooncake connector support GQA transport
#2947 opened Sep 16, 2025 by lidenghui1110 Loading…
3
[Feature] Support moe multi-stream for aclgraph. documentation Improvements or additions to documentation module:core module:ops module:tests ready read for review ready-for-test start test by label for PR
#2946 opened Sep 16, 2025 by whx-sjtu Loading…
8
add test_patch_distributed.py module:tests
#2944 opened Sep 16, 2025 by yangqinghao-cmss Loading…
4
COMM_QUANT_MODE module:core
#2936 opened Sep 15, 2025 by yyoean Loading…
6
[Feat] Add a new kind of linear operation: LayerShardLinear
#2931 opened Sep 15, 2025 by clrs97 Loading…
17
ProTip! Add no:assignee to see everything that’s not assigned.