[Doc][ReleaseNote] Add release notes for v0.16.0rc1 by gcanlin · Pull Request #7067 · vllm-project/vllm-ascend

gcanlin · 2026-03-09T03:33:06Z

Add release notes for v0.16.0rc1

vLLM version: v0.16.0
vLLM main: vllm-project/vllm@4034c3d

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gemini-code-assist · 2026-03-09T03:33:17Z

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces the comprehensive release notes for vLLM Ascend v0.16.0rc1. It documents significant advancements in model support, quantization capabilities, performance optimizations, and hardware compatibility, providing a clear overview of the changes and improvements in this release candidate.

Highlights

Model Support: DeepSeek V3.2 now supports graph mode (piecewise and full_decode_only) with PCP and DCP context parallel, including MTP and chunked prefill.
Model Support: Qwen3-Next now supports PCP and DCP context parallel.
Quantization: 310P now supports W8A8S quantization and saving W8A8SC state.
Quantization: MXFP8 MoE quantization is now supported for Qwen MoE models.
Quantization: Qwen3-Omni quantization adaptation and optimization is now available.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

docs/source/user_guide/release_notes.md
- Added a new section for v0.16.0rc1 release notes.
- Included detailed highlights, features, hardware and operator support, performance improvements, and dependency updates.
- Documented deprecation and breaking changes, as well as various bug fixes for the new release candidate.

Activity

No specific review activity or comments have been recorded for this pull request.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

github-actions · 2026-03-09T03:33:18Z

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

A PR should do only one thing, smaller PRs enable faster reviews.
Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

gemini-code-assist

Code Review

This pull request adds the release notes for version v0.16.0rc1. The changes are limited to documentation. I have no specific comments on the file changes as per the review criteria, which require issues to be of high or critical severity, and documentation issues are typically of lower severity. However, I noticed that the pull request description does not follow the format specified in the repository's style guide (lines 12-39). Please consider updating the description to match the provided template for consistency.

_{Note: Security Review has been skipped due to the limited scope of the PR.}

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

wangxiyuan · 2026-03-09T12:08:08Z

+
+- DeepSeek V3.2 now supports graph mode (piecewise and full_decode_only) with PCP and DCP context parallel. Additionally, PCP now supports MTP and chunked prefill for DeepSeek V3.2. [#6940](https://github.com/vllm-project/vllm-ascend/pull/6940) [#6917](https://github.com/vllm-project/vllm-ascend/pull/6917)
+- Qwen3-Next now supports PCP and DCP context parallel. [#6091](https://github.com/vllm-project/vllm-ascend/pull/6091)
+- MXFP8 MoE quantization is now supported for Qwen MoE models. [#6670](https://github.com/vllm-project/vllm-ascend/pull/6670)


this can be removed, it's only for 950

Done. cc @MengqingCao

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin · 2026-03-10T09:26:48Z

+### Deprecation & Breaking Changes
+
+- `enable_flash_comm_v1` config option has been renamed back to `enable_sp`. [#6883](https://github.com/vllm-project/vllm-ascend/pull/6883)
+- Reverted auto-detect quantization format from model files feature. [#6873](https://github.com/vllm-project/vllm-ascend/pull/6873)


Will remove this when #7111 is merged.

It seems #7111 have some issues with CI, I think we'd better to include it back in the next version to make sure of the quality

MengqingCao · 2026-03-09T03:49:09Z

+
+- DeepSeek V3.2 now supports graph mode (piecewise and full_decode_only) with PCP and DCP context parallel. Additionally, PCP now supports MTP and chunked prefill for DeepSeek V3.2. [#6940](https://github.com/vllm-project/vllm-ascend/pull/6940) [#6917](https://github.com/vllm-project/vllm-ascend/pull/6917)
+- Qwen3-Next now supports PCP and DCP context parallel. [#6091](https://github.com/vllm-project/vllm-ascend/pull/6091)
+- 310P now supports W8A8S quantization and saving W8A8SC state. [#6878](https://github.com/vllm-project/vllm-ascend/pull/6878)


Suggested change

- 310P now supports W8A8S quantization and saving W8A8SC state. [#6878](https://github.com/vllm-project/vllm-ascend/pull/6878)

MengqingCao · 2026-03-09T03:50:41Z

+- fused_qkvzba_split_reshape now supports token number greater than 65536, removing the previous limitation. [#6740](https://github.com/vllm-project/vllm-ascend/pull/6740)
+- NPUWorker Profiler now supports profile_prefix for better profiling experience. [#6968](https://github.com/vllm-project/vllm-ascend/pull/6968)
+- EPLB profiling now displays expert hotness comparison and time required for eplb adjustment. [#6877](https://github.com/vllm-project/vllm-ascend/pull/6877) [#7001](https://github.com/vllm-project/vllm-ascend/pull/7001)
+- Adapt to RecomputeScheduler in vLLM 0.16.0. [#6898](https://github.com/vllm-project/vllm-ascend/pull/6898)


Suggested change

- Adapt to RecomputeScheduler in vLLM 0.16.0. [#6898](https://github.com/vllm-project/vllm-ascend/pull/6898)

MengqingCao · 2026-03-10T14:09:35Z

+
+### Hardware and Operator Support
+
+- 310P now supports w8a8sc quantization method. [#7075](https://github.com/vllm-project/vllm-ascend/pull/7075)


I think we'd better don't metion 310p progress here

MengqingCao · 2026-03-10T14:16:04Z

+### Deprecation & Breaking Changes
+
+- `enable_flash_comm_v1` config option has been renamed back to `enable_sp`. [#6883](https://github.com/vllm-project/vllm-ascend/pull/6883)
+- Reverted auto-detect quantization format from model files feature. [#6873](https://github.com/vllm-project/vllm-ascend/pull/6873)


It seems #7111 have some issues with CI, I think we'd better to include it back in the next version to make sure of the quality

Co-authored-by: Mengqing Cao <cmq0113@163.com> Signed-off-by: Canlin Guo <961750412@qq.com>

MengqingCao

LGTM, thanks for this!

…to qwen3next_graph * 'main' of https://github.com/vllm-project/vllm-ascend: (88 commits) [main][bugfix] Fixed the problem of speculative decoding in FULL mode (vllm-project#7148) fixed fia pad logic in graph mode. (vllm-project#7144) [Doc] fix DSV3.1 PD configs (vllm-project#7187) refactor: add a check before layer_sharding logging (vllm-project#7186) [Build] Add support for Ascend950 chip (vllm-project#7151) Revert "[CI] fix skiped e2e test when upgrade vllm version (vllm-project#6654)" (vllm-project#7166) [MODELRUNNERV2]fix penality ops (vllm-project#7013) [Bugfix][LoRA] Fix the issue when enable LoRA + tp + fully_sharded_loras (vllm-project#6650) [KV Pool]get_num_new_matched_tokens return 0 if token length < block_size (vllm-project#7146) [CI] Build Image for v0.16.0rc1 (vllm-project#7155) [CI] Skip `test_mooncake_layerwise_connector.py` in `ut` (vllm-project#7147) [BugFix]Fix recomputed scheduler bug (vllm-project#7137) [Model] Support Minimax-m2.5 on NPU (vllm-project#7105) [P/D]Mooncake Layerwise Connector supports hybrid attention manager with multiple kvcache groups (vllm-project#7022) Add patch_qwen3_5 for triton ops fused_recurrent_gated_delta_rule (vllm-project#7109) [Doc][ReleaseNote] Add release notes for v0.16.0rc1 (vllm-project#7067) [Misc] Download on both hk and guiyang region (vllm-project#7129) [bugdix] The problem that the w4a8 weight fails to be loaded when the EP is not enabled is resolved. (vllm-project#7090) [eagle][cp] fix eagle_cp enable bug2 (vllm-project#7079) [CI]Upgrade niglty multi-node-tests max-parallel to 2 (vllm-project#7035) ...

Add release notes for v0.16.0rc1 - vLLM version: v0.16.0 - vLLM main: vllm-project/vllm@4034c3d --------- Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: Canlin Guo <961750412@qq.com> Co-authored-by: Mengqing Cao <cmq0113@163.com>

[Doc][Misc] Add release notes for v0.16.0rc1

e9d2d42

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin requested review from LCAIZJ, Yikun and wangxiyuan as code owners March 9, 2026 03:33

github-actions bot added the documentation Improvements or additions to documentation label Mar 9, 2026

gemini-code-assist bot reviewed Mar 9, 2026

View reviewed changes

gcanlin commented Mar 9, 2026

View reviewed changes

update

e683369

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

wangxiyuan reviewed Mar 9, 2026

View reviewed changes

gcanlin added 7 commits March 10, 2026 08:31

change

d585b31

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

Merge branch 'main' into add_016_releasenote

33ef69d

Merge branch 'main' into add_016_releasenote

faa1d77

update

4006d94

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

update

56f93af

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

remove a5

7412ea3

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

fix lint

3f9e15e

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin commented Mar 10, 2026

View reviewed changes

MengqingCao changed the title ~~[Doc][Misc] Add release notes for v0.16.0rc1~~ [Doc][ReleaseNote] Add release notes for v0.16.0rc1 Mar 10, 2026

MengqingCao reviewed Mar 10, 2026

View reviewed changes

gcanlin and others added 2 commits March 10, 2026 22:35

Apply suggestions from code review

a90e1bc

Co-authored-by: Mengqing Cao <cmq0113@163.com> Signed-off-by: Canlin Guo <961750412@qq.com>

Update docs/source/user_guide/release_notes.md

f8676af

Co-authored-by: Mengqing Cao <cmq0113@163.com> Signed-off-by: Canlin Guo <961750412@qq.com>

MengqingCao approved these changes Mar 10, 2026

View reviewed changes

MengqingCao merged commit a78a00e into vllm-project:main Mar 10, 2026
22 checks passed


		### Hardware and Operator Support

		- 310P now supports w8a8sc quantization method. [#7075](https://github.com/vllm-project/vllm-ascend/pull/7075)

Conversation

gcanlin commented Mar 9, 2026 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

gemini-code-assist bot commented Mar 9, 2026

Summary of Changes

Highlights

Footnotes

Uh oh!

github-actions bot commented Mar 9, 2026

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

MengqingCao left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

gcanlin commented Mar 9, 2026 •

edited by github-actions bot

Loading