Skip to content

Pull requests: vllm-project/vllm-ascend

New pull request New
100 Open 843 Closed
100 Open 843 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

Add profiling multimodal model step and fix the OOM bug when profilin…
#1408 opened Jun 24, 2025 by ChenTaoyu-SJTU Loading…
1
rm router logits Improve TTOP 3ms module:core module:ops
#1407 opened Jun 24, 2025 by ttanzhiqiang Loading…
1
[V0.9.1] Prevent Forced Stream Synchronization Triggered by Environme… module:core
#1405 opened Jun 24, 2025 by rjg-lyh Loading…
[BugFix] Fix the problem that torchair doesn't support tp > 4.
#1404 opened Jun 24, 2025 by whx-sjtu Loading…
[Fix] Prevent Forced Stream Synchronization Triggered by Environment … module:core
#1403 opened Jun 24, 2025 by rjg-lyh Loading…
2
V0.9.1 dev module:ops module:quantization
#1402 opened Jun 24, 2025 by zhenwenqi2024 Loading…
[Enhancement] Enable MODELSCOPE in tests
#1400 opened Jun 24, 2025 by yiz-liu Loading…
[Bugfix] Fix GatherV3 bug
#1397 opened Jun 24, 2025 by yiz-liu Loading…
3
shared_experts+router_experts merge all_reduce(Improve TTOP 5ms) module:core module:ops ready read for review
#1395 opened Jun 24, 2025 by ttanzhiqiang Loading…
11
[Doc] Add Qwen2.5-VL eager mode doc documentation Improvements or additions to documentation
#1394 opened Jun 24, 2025 by shen-shanshan Loading…
[Refactor] Remove duplicate multimodal codes in ModelRunner
#1393 opened Jun 24, 2025 by yiz-liu Loading…
[Doc] Add performance tuning doc to main documentation Improvements or additions to documentation
#1392 opened Jun 24, 2025 by shen-shanshan Loading…
【Feature】Dynamic Expert Load Balance Zero-like-overhead merge-conflicts module:core module:ops module:quantization
#1391 opened Jun 24, 2025 by raindaywhu Loading…
1
[Feature]Moe alltoallv communication optimization for unquantized RL training sence & alltoallv support dpo merge-conflicts module:core module:ops
#1389 opened Jun 24, 2025 by harygo22 Loading…
2
[Build] Add build info module:core module:ops module:tests
#1386 opened Jun 24, 2025 by wangxiyuan Loading…
1
[BugFix]Remove not using patch_eagle.py for CI. ready read for review
#1385 opened Jun 24, 2025 by yuancaoyaoHW Loading…
4
[WIP][ExternalDP][RL] Make external DP support on EP and ETP module:tests
#1384 opened Jun 24, 2025 by MengqingCao • Draft
1 task
1
[Bugfix] Support Qwen3-MOE on aclgraph mode module:ops
#1381 opened Jun 23, 2025 by ApsarasX Loading…
2
[Bugfix] Fix memory-leak caused by dist._functional_collectives.reduce_scatter_tensor module:ops ready read for review
#1380 opened Jun 23, 2025 by ApsarasX Loading…
2
[BugFix] Fix a bug of running chunked-prefill with torchair.
#1378 opened Jun 23, 2025 by whx-sjtu Loading…
1
[WIP]FC3 merge-conflicts module:ops module:quantization
#1377 opened Jun 23, 2025 by nakairika Loading…
1
[Bugfix] Fix sleep mode level 2
#1376 opened Jun 23, 2025 by Potabk Loading…
1
Doc Enhancement: Single NPU(Qwen3-8B) aclgraph mode + eager mode documentation Improvements or additions to documentation
#1374 opened Jun 23, 2025 by leo-pony Loading…
[Misc] Clean up uesless code for LLM initialize
#1373 opened Jun 23, 2025 by wangxiyuan Loading…
4
[Doc] Add qwen2-audio eager mode tutorial documentation Improvements or additions to documentation
#1371 opened Jun 23, 2025 by shen-shanshan Loading…
1
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.