fca2f948c1
### What this PR does / why we need it?
The test case `tests/e2e/multicard/test_data_parallel.py` was skipped
due to the errors encountered during migration from Ascend A2 to A3, the
details are as follows
```
(EngineCore_DP0 pid=17833) RuntimeError: npu_moe_distribute_dispatch_v2:build/CMakeFiles/torch_npu.dir/compiler_depend.ts:161 NPU function error: call aclnnMoeDistributeDispatchV3 failed, error code is 561002
(EngineCore_DP0 pid=17833) [ERROR] 2025-12-23-07:36:19 (PID:17833, Device:0, RankID:-1) ERR00100 PTA call acl api failed.
(EngineCore_DP0 pid=17833) EZ9999: Inner Error!
(EngineCore_DP0 pid=17833) EZ9999[PID: 17833] 2025-12-23-07:36:19.237.396 (EZ9999): HCCL_BUFFSIZE is too SMALL, maxBs = 512, h = 2048, epWorldSize = 2, localMoeExpertNum = 64, sharedExpertNum = 0, tokenNeedSizeDispatch = 4608, tokenNeedSizeCombine = 4096, k = 8, NEEDED_HCCL_BUFFSIZE(((maxBs * tokenNeedSizeDispatch * ep_worldsize * localMoeExpertNum) + (maxBs * tokenNeedSizeCombine * (k + sharedExpertNum))) * 2) = 609MB, HCCL_BUFFSIZE=200MB.[FUNC:MoeDistributeDispatchA3TilingFuncImpl][FILE:moe_distribute_dispatch_v2_tiling.cc][LINE:941]
(EngineCore_DP0 pid=17833) TraceBack (most recent call last):
(EngineCore_DP0 pid=17833) MoeDistributeDispatchV2 do tiling failed, ret is -1.
(EngineCore_DP0 pid=17833) Check NnopbaseExecutorDoTiling(executor) failed
(EngineCore_DP0 pid=17833) Check NnopbaseExecutorTilingAndUpdateBinInfo(executor) failed
(EngineCore_DP0 pid=17833) Check NnopbaseExecutorMatchCache(executor) failed
(EngineCore_DP0 pid=17833) Check NnopbaseRunForWorkspace(*executor, workspaceSize) failed
```
### Does this PR introduce _any_ user-facing change?
None
### How was this patch tested?
After fixed, I ran `pytest -sv --durations=0
tests/e2e/multicard/test_data_parallel.py`, and the result looks good
```
========================================================================================= warnings summary =========================================================================================
<frozen importlib._bootstrap>:241
<frozen importlib._bootstrap>:241: DeprecationWarning: builtin type SwigPyPacked has no __module__ attribute
<frozen importlib._bootstrap>:241
<frozen importlib._bootstrap>:241: DeprecationWarning: builtin type SwigPyObject has no __module__ attribute
-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
======================================================================================== slowest durations =========================================================================================
112.69s call tests/e2e/multicard/test_data_parallel.py::test_qwen_inference_dp2[32-vllm-ascend/Qwen3-30B-A3B-W8A8]
88.11s call tests/e2e/multicard/test_data_parallel.py::test_qwen_inference_dp2[32-Qwen/Qwen3-30B-A3B]
70.06s call tests/e2e/multicard/test_data_parallel.py::test_qwen_inference_dp2[32-Qwen/Qwen3-0.6B]
(6 durations < 0.005s hidden. Use -vv to show these durations.)
============================================================================ 3 passed, 2 warnings in 270.88s (0:04:30) ============================================================================
```
- vLLM version: release/v0.13.0
- vLLM main:
ad32e3e19c
Signed-off-by: wjunLu <wjunlu217@gmail.com>
vLLM Ascend Plugin
| About Ascend | Documentation | #sig-ascend | Users Forum | Weekly Meeting |
English | 中文
Latest News 🔥
- [2025/12] We released the new official version v0.11.0! Please follow the official guide to start using vLLM Ascend Plugin on Ascend.
- [2025/09] We released the new official version v0.9.1! Please follow the official guide to start deploy large scale Expert Parallelism (EP) on Ascend.
- [2025/08] We hosted the vLLM Beijing Meetup with vLLM and Tencent! Please find the meetup slides here.
- [2025/06] User stories page is now live! It kicks off with LLaMA-Factory/verl//TRL/GPUStack to demonstrate how vLLM Ascend assists Ascend users in enhancing their experience across fine-tuning, evaluation, reinforcement learning (RL), and deployment scenarios.
- [2025/06] Contributors page is now live! All contributions deserve to be recorded, thanks for all contributors.
- [2025/05] We've released first official version v0.7.3! We collaborated with the vLLM community to publish a blog post sharing our practice: Introducing vLLM Hardware Plugin, Best Practice from Ascend NPU.
- [2025/03] We hosted the vLLM Beijing Meetup with vLLM team! Please find the meetup slides here.
- [2025/02] vLLM community officially created vllm-project/vllm-ascend repo for running vLLM seamlessly on the Ascend NPU.
- [2024/12] We are working with the vLLM community to support [RFC]: Hardware pluggable.
Overview
vLLM Ascend (vllm-ascend) is a community maintained hardware plugin for running vLLM seamlessly on the Ascend NPU.
It is the recommended approach for supporting the Ascend backend within the vLLM community. It adheres to the principles outlined in the [RFC]: Hardware pluggable, providing a hardware-pluggable interface that decouples the integration of the Ascend NPU with vLLM.
By using vLLM Ascend plugin, popular open-source models, including Transformer-like, Mixture-of-Expert, Embedding, Multi-modal LLMs can run seamlessly on the Ascend NPU.
Prerequisites
- Hardware: Atlas 800I A2 Inference series, Atlas A2 Training series, Atlas 800I A3 Inference series, Atlas A3 Training series, Atlas 300I Duo (Experimental)
- OS: Linux
- Software:
- Python >= 3.10, < 3.12
- CANN == 8.3.rc2 (Ascend HDK version refers to here)
- PyTorch == 2.8.0, torch-npu == 2.8.0
- vLLM (the same version as vllm-ascend)
Getting Started
Please use the following recommended versions to get started quickly:
| Version | Release type | Doc |
|---|---|---|
| v0.12.0rc1 | Latest release candidate | QuickStart and Installation for more details |
| v0.11.0 | Latest stable version | QuickStart and Installation for more details |
Contributing
See CONTRIBUTING for more details, which is a step-by-step guide to help you set up development environment, build and test.
We welcome and value any contributions and collaborations:
- Please let us know if you encounter a bug by filing an issue
- Please use User forum for usage questions and help.
Branch
vllm-ascend has main branch and dev branch.
- main: main branch，corresponds to the vLLM main branch, and is continuously monitored for quality through Ascend CI.
- vX.Y.Z-dev: development branch, created with part of new releases of vLLM. For example,
v0.7.3-devis the dev branch for vLLMv0.7.3version.
Below is maintained branches:
| Branch | Status | Note |
|---|---|---|
| main | Maintained | CI commitment for vLLM main branch and vLLM v0.13.0 tag |
| v0.7.1-dev | Unmaintained | Only doc fixed is allowed |
| v0.7.3-dev | Maintained | CI commitment for vLLM 0.7.3 version, only bug fix is allowed and no new release tag any more. |
| v0.9.1-dev | Maintained | CI commitment for vLLM 0.9.1 version |
| v0.11.0-dev | Maintained | CI commitment for vLLM 0.11.0 version |
| rfc/feature-name | Maintained | Feature branches for collaboration |
Please refer to Versioning policy for more details.
Weekly Meeting
- vLLM Ascend Weekly Meeting: https://tinyurl.com/vllm-ascend-meeting
- Wednesday, 15:00 - 16:00 (UTC+8, Convert to your timezone)
License
Apache License 2.0, as found in the LICENSE file.