Go to file
Hank b78dab8bb7
test / Detect Changed Paths (push) Successful in 1m11s Details
test / e2e_test (vllm-metax-runner-set, main) (push) Has been cancelled Details
test / e2e_test (vllm-metax-runner-set, v0.10.2) (push) Has been cancelled Details
[doc] update docs
Signed-off-by: Hank <hcc.mayday@gmail.com>
2026-06-10 08:12:19 +00:00
.gemini fix: workflow error 2025-09-12 10:59:04 +08:00
.github chore: fix dailytest failure (#102) 2025-10-15 13:43:49 +08:00
.modelpath [test] add lmcache offloading test(#101) 2025-10-23 09:58:43 +08:00
cmake [chore] fix compile error 2026-06-02 07:20:45 +00:00
csrc [bugfix] shared memory exceeded in maca backend 2026-05-15 18:05:30 +08:00
docker [dockerfile] update docker file 2026-05-13 16:20:28 +08:00
docs [doc] update docs 2026-06-10 08:12:19 +00:00
examples [update] sync examples with upstream 2026-05-25 10:16:24 +08:00
requirements [version] synchronize to v0.22.0's requirments 2026-05-29 06:33:24 +00:00
tests Support vllm 0.17.0 (#216) 2026-03-17 22:15:06 +08:00
tools [pre-commit] fix typos 2026-04-27 11:24:48 +08:00
vllm_metax [BugFix] correct `cu_seqlens_k` in dcp when capture cudagraph 2026-06-10 08:06:19 +00:00
.clang-format chore: add pre-commit hooks 2025-09-09 13:20:51 +08:00
.dockerignore DockerFile: support docker file on ubi:9.6 2025-10-29 11:10:49 +08:00
.gitignore chore: add pre-commit workflow 2025-09-09 18:10:28 +08:00
.markdownlint.yaml feat!: support v0.11.1 (#112) 2025-11-14 10:28:40 +08:00
.pre-commit-config.yaml feat!: support v0.11.1 (#112) 2025-11-14 10:28:40 +08:00
.readthedocs.yaml doc: fix builds error 2025-11-05 14:29:05 +08:00
.shellcheckrc feat!: support v0.11.1 (#112) 2025-11-14 10:28:40 +08:00
.yapfignore feat!: support v0.11.1 (#112) 2025-11-14 10:28:40 +08:00
CMakeLists.txt [DeepSeek_v4] Rebase deepseek v4 (#257) 2026-05-15 17:47:43 +08:00
CODE_OF_CONDUCT.md init 2025-06-19 07:01:43 +00:00
CONTRIBUTING.md Update CONTRIBUTING.md 2025-09-09 14:32:22 +08:00
DCO init 2025-06-19 07:01:43 +00:00
LICENSE init 2025-06-19 07:01:43 +00:00
MANIFEST.in [Bugfix][Build] FusedMoE configs folders failed to be included 2026-04-08 14:40:58 +08:00
README.md [doc] update docs 2026-06-10 08:12:19 +00:00
env.sh Support vllm 0.15.0 (#209) 2026-02-26 18:18:55 +08:00
format.sh [chores] remove +x privilege 2026-05-13 16:20:28 +08:00
mkdocs.yaml doc: remove unnecessary widgets 2025-11-07 14:07:14 +08:00
pyproject.toml [pre-commit] fix typos 2026-04-27 11:24:48 +08:00
setup.py [version] synchronize to v0.22.0's requirments 2026-05-29 06:33:24 +00:00
use_existing_metax.py Support vllm 0.17.0 (#216) 2026-03-17 22:15:06 +08:00

README.md

vLLM

vLLM MetaX Plugin

DeepWiki

| About MetaX | Documentation | #sig-maca |


Latest News 🔥

  • [2026/6] Released vllm-metax v0.20.0 🚀 — aligned with vLLM v0.20.0.
  • [2026/5] Released vllm-metax v0.19.0 🎉 — aligned with vLLM v0.19.0, but fully supported Gemma4!
  • [2026/4] Released vllm-metax v0.18.0 😎 — aligned with vLLM v0.18.0, same as usual.
  • [2026/3] Released vllm-metax v0.17.0 🎉 — aligned with vLLM v0.17.0, supported more models and improved performance.
  • [2026/3] Released vllm-metax v0.15.0 🦐 — aligned with vLLM v0.15.0, more models and more features!
  • [2026/3] Released vllm-metax v0.14.0 🚀 — aligned with vLLM v0.14.0, same as usual.
  • [2026/2] Released vllm-metax v0.13.0 🧨 — aligned with vLLM v0.13.0, brings you the latest features and model in v0.13.0!
  • [2026/1] Released vllm-metax v0.12.0 😎 — aligned with vLLM v0.12.0, supported more models and improved performance.
  • [2026/1] Released vllm-metax v0.11.2 👻 — aligned with vLLM v0.11.2, supported more models and improved performance.
  • [2025/11] Released vllm-metax v0.10.2 🎉 — aligned with vLLM v0.10.2, improved model performance, and fixed key decoding bugs.
  • [2025/11] We hosted vLLM Beijing Meetup focusing on distributed inference and diverse accelerator support with vLLM! Please find the meetup slides here.
  • [2025/08] We hosted vLLM Shanghai Meetup focusing on building, developing, and integrating with vLLM! Please find the meetup slides here.

About

vLLM MetaX is a hardware plugin that enables vLLM to run seamlessly on MetaX GPUs. MetaX provides a cuda-like backend through MACA, delivering a near-native CUDA experience on MetaX hardware.

It is the recommended approach for supporting the MetaX backend within the vLLM community.

The plugin is implemented in accordance with the vLLM plugin RFCs:

These RFCs help ensure proper feature and functionality support when integrating MetaX GPUs with vLLM.

Prerequisites

  • Hardware: MetaX C-series
  • OS: Linux
  • Software:
    • Python >= 3.10, <= 3.12
    • vLLM (the same version as vllm-metax)
    • Docker support

Getting Started

vLLM MetaX currently supports deployment only with Docker images released by the MetaX developer community, which work out of the box.

The Dockerfile for other OS environments is still under testing.

If you want to develop, debug, or test the latest features in vllm-metax, you may need to build it from source. Please follow this source build tutorial.

Branch

vllm-metax has three kinds of branches.

  • master: the main branch, which tracks the upstream vLLM main branch.
  • vX.Y.Z-dev: development branches created after a vLLM release.

    For example, v0.1x.0-dev is the development branch for a newly released branch like releases/v0.1x.0.

  • releases/vX.Y.Z: release branches created from v0.1x.0-dev, indicating that the corresponding vllm-metax development branch has been fully tested and released.

    For example, vllm-metax's releases/v0.1x.0 corresponds to vLLM's releases/v0.1x.0. The same naming rule applies to tags.

Below are the maintained branches:

Branch Status Note
master N/A Tracks vLLM main; functionality is not guaranteed
releases/v0.20.0 Released Corresponds to vLLM release v0.20.0
releases/v0.19.0 Released Corresponds to vLLM release v0.19.0
releases/v0.18.0 Released Corresponds to vLLM release v0.18.0
releases/v0.17.0 Released Corresponds to vLLM release v0.17.0
v0.16.0 N/A Skipped
releases/v0.15.0 Released Corresponds to vLLM release v0.15.0
releases/v0.14.0 Released Corresponds to vLLM release v0.14.0
releases/v0.13.0 Released Corresponds to vLLM release v0.13.0
releases/v0.12.0 Released Corresponds to vLLM release v0.12.0
releases/v0.11.2 Released Corresponds to vLLM release v0.11.2
releases/v0.10.2 Released Corresponds to vLLM release v0.10.2

For more details, please check the Quickstart Guide.

License

Apache License 2.0, as found in the LICENSE file.