Skip to content

Pull requests: vllm-project/vllm-ascend

New pull request New
198 Open 2,745 Closed
198 Open 2,745 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

[0.11.0][HybridKV] Support KV sharing in mambaspec and fullattnspec
#4210 opened Nov 14, 2025 by MengqingCao Loading…
3
[long seq feat]GQA support long-prefill-token-threshold
#4209 opened Nov 14, 2025 by Delphine-Nic Loading…
2
[v0.11.0-dev][misc]change default capture size for Qwen3-MoE when using full dp module:core
#4205 opened Nov 14, 2025 by Angazenn Loading…
2
[Cherry-pick][0.11.0] Adapted to torch_npu.npu_fused_infer_attention_score ready read for review ready-for-test start test by label for PR
#4202 opened Nov 14, 2025 by wxsIcey Loading…
4
[main][misc]change default capture size for Qwen3-MoE when using pure dp module:core module:tests ready read for review ready-for-test start test by label for PR
#4199 opened Nov 14, 2025 by Angazenn Loading…
3
[MM][Perf] Replace VisionPatchEmbed with that in vllm for better performance
#4198 opened Nov 14, 2025 by shen-shanshan Loading…
2
[HybridKV] Support KV sharing in mambaspec and fullattnspec module:tests ready read for review ready-for-test start test by label for PR
#4196 opened Nov 14, 2025 by MengqingCao Loading…
1
1
[Kernel] add custom moe ops for prefill
#4194 opened Nov 14, 2025 by shiro-zzzz Loading…
8
optimiation encoder attention
#4192 opened Nov 14, 2025 by Jeaniowang Loading…
2
[Test] enhance test coverage for model_runner_v1 module:tests
#4189 opened Nov 14, 2025 by noemotiovon Loading…
4
[Feat] Flashcomm2 use o_shared linear module:core module:tests
#4188 opened Nov 13, 2025 by zzhx1 Loading…
5
test-lora-310p-build
#4187 opened Nov 13, 2025 by liuchenbing Loading…
6
revise doc documentation Improvements or additions to documentation
#4185 opened Nov 13, 2025 by Angazenn Loading…
3
[bugfix] add mlapo test script module:tests ready read for review ready-for-test start test by label for PR
#4184 opened Nov 13, 2025 by chenjunyi-dev Loading…
2
[feature] Mooncake_connector support pcp/dcp
#4183 opened Nov 13, 2025 by wangxiaochao6 Loading…
9
[P/D] Add readme for PD separation documentation Improvements or additions to documentation
#4182 opened Nov 13, 2025 by wangxiaoteng888 Loading…
8
make vllm-ascend work well in developer mode module:core
#4179 opened Nov 13, 2025 by Ronald1995 Loading…
6
Encoder separation for Encode-Prefill-Decode Disaggregation
#4176 opened Nov 13, 2025 by amy-why-3459 Loading…
4
Adopt inductor fusion and define quantization fusion pass merge-conflicts module:core module:ops
#4168 opened Nov 13, 2025 by wxsIcey Loading…
11
[Bug_fix] fix torchair o_proj forward parameter module:ops
#4166 opened Nov 13, 2025 by zzhx1 Loading…
4
qwen3-vl Vit module enable sp and mrope fusion op module:multimodal
#4165 opened Nov 13, 2025 by qigangc Loading…
4
[DoNotMerge]fia pa documentation Improvements or additions to documentation module:tests
#4163 opened Nov 13, 2025 by Angazenn Loading…
4
[P/D] pd proxy support ipv6 documentation Improvements or additions to documentation
#4161 opened Nov 13, 2025 by liziyu179 Loading…
6
[ops] npu_top_k_top_p supports k or p is None
#4154 opened Nov 12, 2025 by linfeng-yuan Loading…
3
[0.11.0][ops] npu_top_k_top_p supports k and p only
#4153 opened Nov 12, 2025 by linfeng-yuan Loading…
3
ProTip! Mix and match filters to narrow down what you’re looking for.