EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Doc][Misc] Update release notes for v0.15.0rc1 (#6859)

Browse Source 
### What this PR does / why we need it?

This PR updates the release notes for `v0.15.0rc1` to:
- Mark the `310P MoE and W8A8 Support` feature as experimental.
- Add a note for `Kimi-K2.5 Model Support` clarifying that it has known
issues in vLLM 0.15.0 and requires manual patching to work correctly.

### Does this PR introduce _any_ user-facing change?

No, this is a documentation-only update.

### How was this patch tested?

N/A (documentation change).

- vLLM version: v0.16.0
- vLLM main:
15d76f74e2

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2026-02-27 22:35:09 +08:00
committed by GitHub
parent b60b991005
commit 9cd0d6c33d
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/source/user_guide/release_notes.md
View File

@@ -7,9 +7,9 @@ This is the first release candidate of v0.15.0 for vLLM Ascend. Please follow th
### Highlights
- **NPU Graph EX (npugraph_ex) Enabled by Default**: The npugraph_ex feature is now enabled by default, providing better graph optimization with integrated inductor pass and MatmulAllReduceAddRMSNorm fusion. [#6354](https://github.com/vllm-project/vllm-ascend/pull/6354) [#6664](https://github.com/vllm-project/vllm-ascend/pull/6664) [#6006](https://github.com/vllm-project/vllm-ascend/pull/6006)
- **310P MoE and W8A8 Support**: 310P now supports MoE models, W8A8 quantization, and weightNZ feature, significantly expanding hardware capabilities. [#6530](https://github.com/vllm-project/vllm-ascend/pull/6530) [#6641](https://github.com/vllm-project/vllm-ascend/pull/6641) [#6454](https://github.com/vllm-project/vllm-ascend/pull/6454) [#6705](https://github.com/vllm-project/vllm-ascend/pull/6705)
- **310P MoE and W8A8 Support**[Experimental]: 310P now supports MoE models, W8A8 quantization, and weightNZ feature, significantly expanding hardware capabilities. [#6530](https://github.com/vllm-project/vllm-ascend/pull/6530) [#6641](https://github.com/vllm-project/vllm-ascend/pull/6641) [#6454](https://github.com/vllm-project/vllm-ascend/pull/6454) [#6705](https://github.com/vllm-project/vllm-ascend/pull/6705)
- **Qwen3-VL-MoE EAGLE Support**: Added EAGLE speculative decoding support for Qwen3-VL-MoE model. [#6327](https://github.com/vllm-project/vllm-ascend/pull/6327)
- **Kimi-K2.5 Model Support**: Added support for Kimi-K2.5 models. [#6755](https://github.com/vllm-project/vllm-ascend/pull/6755)
- **Kimi-K2.5 Model Support**: Added support for Kimi-K2.5 models. **Please note** that vLLM 0.15.0 has a known issue with Kimi-K2.5. To fix this, please apply the changes from the upstream `vllm-project/vllm` repository, specifically from pull requests [#33320](https://github.com/vllm-project/vllm/pull/33320) and [#34501](https://github.com/vllm-project/vllm/pull/34501). [#6755](https://github.com/vllm-project/vllm-ascend/pull/6755)
### Features