Project-HAMi /HAMi

Heterogeneous GPU Sharing on Kubernetes

★ 5,014⑂ 894👁 28⑃ 2,152● 1,269▤ 18
GoApache-2.0Created 2021-09-14project-hami.io
Download: v2.11.0v2.11.0 ↓2.11.02026-11-04
+4
Daily
5,014
stars
894
forks
28
watchers
2,152
prs
1,269
issues
18
releases

README

English version | 中文版 | 日本語版
















 HAMi

Kubernetes GPU virtualization and heterogeneous accelerator scheduling for AI infrastructure.

!HAMi Architecture

HAMi stands for Heterogeneous AI Computing Virtualization Middleware. Formerly known as k8s-vGPU-scheduler, HAMi helps platform teams share expensive GPUs and other AI accelerators across Kubernetes workloads, isolate device memory and compute, and schedule pods with device-aware policies without changing application code.

HAMi is a CNCF Incubating and CNCF Landscape project. It is also listed in the CNAI Landscape.

!CNCF logo

 Documentation

- Architecture
- GPU virtualization
- Supported devices and vendor guides
- Prerequisites
- Install and run a first workload
- Upgrade and migrate values
- Configuration and operations
- NVIDIA scheduling policies
- NVIDIA dynamic MIG
- Device allocation metrics
- Device usage metrics
- Grafana dashboards
- WebUI
- Volcano integration
- Kueue integration
- Benchmarks
- Troubleshooting
- Uninstall

 Repository resources

- Chart values reference
- Workload examples
- Agent skills
- HAMi-WebUI repository

 Roadmap, Governance, And Contributing

HAMi is governed by maintainers and contributors. Governance is described in the HAMi community repository.

To contribute code, documentation, tests, or device backend improvements, read CONTRIBUTING.md.

 Community

For chat channels, mailing lists, meeting schedules, and meeting notes, visit the HAMi community page.

Talk materials

 License

HAMi is licensed under the Apache License 2.0. See LICENSE for details.

Copyright Contributors to HAMi, established as HAMi a Series of LF Projects, LLC.

Top issues

Discussions

Latest release

v2.11.0 v2.11.0 2026-11-04
What's Changed

 Key Features

 General:
No longer mount HAMI-core to whole GPU allocation, metrics will be exported by NVML directly
Flexible-MIG
Support Cluster Autoscaler
Real-usage monitoring for Flexible-mig workloads

 Remote CUDA:
Allows running vllm on remote cuda

 Schedule:
PodGroup Support
New schedule policy: mutex
Schedule policy combinations
GPU/CPU Numa Align

 Devices:
Full AMD Support
Ascend 950 Support
libvnpu performance optimization



 ✨ Features Details
 ci: allow manual dispatch of the Trivy image scan by @moezdil in https://github.com/Project-

Snapshot: 2026-11-16T09:23:14Z · Data license Apache-2.0 · JSON