-
Notifications
You must be signed in to change notification settings - Fork 544
Closed as not planned
Labels
releaserelease relatedrelease related
Description
Release Checklist
Release Version: v0.9.1rc2
Release Branch: main
Release Date: 0628
Release Manager: @Yikun
Prepare Release Note
- [v0.9.1rc2] FAQ / Feedback | 问题/反馈 #1487
- Write the release note PR: Bump version to v0.9.1rc2 #1488
- Update the feedback issue link in docs/source/faqs.md
- Add release note to docs/source/user_guide/release_notes.md
- Update version info in docs/source/community/versioning_policy.md
- Update contributor info in docs/source/community/contributors.md
- Update package version in docs/conf.py
PR need Merge
- [PromptLogprobs][V1] Support prompt logprobs to fix ceval accuracy in V1 #1483
- [PERF]support MERRouter #1421
- [PERF]support H2P communication optimization for PanguProMoe #1463
- support pangumoe w8a8c8 and docs #1477
- [BugFix]Fix bugs when initializing communication groups with dp on 300I Duo #1478
- [Core] Fix block table shape to make Prefix cache work with Ascend scheduler #1446
- [CORE]initial support for torchair with non-mla backend #1506
- Fix W8A8 fused moe bug #1529
- [Bugfix] Support Qwen3-MOE on aclgraph mode #1381
- Fix wheel glibc version incompatibility #1582
Functional Test
-
New model
-
Altlas A2 series 2025.06.21
- Docker image E2E test
- Performance
- Accuracy
- Doc Turtorial:
Multi-NPU (XXXX 72B)@shen-shanshan
-
Altlas 300I DUO series @leo-pony
- Docker image E2E test
- Performance
- Accuracy
- Doc Turtorial:
Multi-NPU (300I DUO)
-
Quantization @Angazenn
- Performance
- Accuracy
-
-
Accuracy report (auto)
-
Performance for Qwen2 / Qwen3 / Qwen2.5 VL
-
DeepSeek test
Doc Test
- Tutorial is updated.
- User Guide is updated.
- Developer Guide is updated.
Prepare Artifacts
- Docker image is ready.
- Wheel package is ready.
Release Step
- Release note PR is merged.
- Post the release on GitHub release page.
- Generate official doc page on https://app.readthedocs.org/dashboard/
- Wait for the wheel package to be available on https://pypi.org/project/vllm-ascend
- Wait for the docker image to be available on https://quay.io/ascend/vllm-ascend
- Upload 310p wheel to Github release page
- Brodcast the release news (By message, blog , etc)
- Close this issue
Metadata
Metadata
Assignees
Labels
releaserelease relatedrelease related