| About MetaX | Documentation | #sig-maca |
Latest News 🔥
- [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.
vLLM MetaX is a hardware plugin for running vLLM seamlessly on MetaX GPU, which is a cuda_alike backend and provided near-native CUDA experiences on MetaX Hardware with MACA.
It is the recommended approach for supporting the MetaX backend within the vLLM community.
The plugin follows the vLLM plugin RFCs by default:
Which ensured the hardware features and functionality support on integration of the MetaX GPU with vLLM.
- Hardware: MetaX C-series
- OS: Linux
- Software:
- Python >= 3.9, < 3.12
- vLLM (the same version as vllm-metax)
- Docker support
vLLM MetaX currently only support starting on docker images release by MetaX develop community which is out of box. (DockerFile for other OS is undertesting)
If you want to develop, debug or test the newest feature on vllm-metax, you may need to build from scratch and follow this source build tutorial.
vllm-metax has master branch and dev branch.
- master: main branch，catching up with main branch of vLLM upstream.
- vX.Y.Z-dev: development branch, created with part of new releases of vLLM. For example,
v0.10.2-devis the dev branch for vLLMv0.10.2version.
Below is maintained branches:
| Branch | Status | Note |
|---|---|---|
| master | Maintained | trying to support vllm main, no gurantee on functionality |
| v0.11.1rc6-dev | Maintained | under testing |
| v0.11.0-dev | Maintained | under testing |
| releases/v0.10.2 | Maintained | Released |
Please check here for more details .
Apache License 2.0, as found in the LICENSE file.
