{"id":25905673,"url":"https://github.com/nexusgpu/tensor-fusion","last_synced_at":"2026-05-24T04:01:00.939Z","repository":{"id":266251195,"uuid":"887587896","full_name":"NexusGPU/tensor-fusion","owner":"NexusGPU","description":"Tensor Fusion is a state-of-the-art GPU virtualization and pooling solution designed to optimize GPU cluster utilization to its fullest potential.","archived":false,"fork":false,"pushed_at":"2026-05-18T04:35:28.000Z","size":4603,"stargazers_count":151,"open_issues_count":15,"forks_count":31,"subscribers_count":2,"default_branch":"main","last_synced_at":"2026-05-18T06:51:49.636Z","etag":null,"topics":["ai","amd-gpu","autoscaling","dynamic-resource-allocation","gpu","gpu-acceleration","gpu-pooling","gpu-scheduling","gpu-usage","gpu-virtualization","inference","karpenter","kubernetes","llm-serving","nvidia","pytorch","rcuda","remote-gpu","vgpu"],"latest_commit_sha":null,"homepage":"https://tensor-fusion.ai","language":"Go","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/NexusGPU.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2024-11-12T23:49:48.000Z","updated_at":"2026-05-18T04:33:21.000Z","dependencies_parsed_at":"2025-01-11T16:19:45.421Z","dependency_job_id":"8fdf43c1-3e14-4857-b568-0ae5684dc2ec","html_url":"https://github.com/NexusGPU/tensor-fusion","commit_stats":null,"previous_names":["nexusgpu/tensor-fusion-operator","nexusgpu/tensor-fusion"],"tags_count":388,"template":false,"template_full_name":null,"purl":"pkg:github/NexusGPU/tensor-fusion","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NexusGPU%2Ftensor-fusion","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NexusGPU%2Ftensor-fusion/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NexusGPU%2Ftensor-fusion/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NexusGPU%2Ftensor-fusion/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/NexusGPU","download_url":"https://codeload.github.com/NexusGPU/tensor-fusion/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NexusGPU%2Ftensor-fusion/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33420660,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-23T22:14:44.296Z","status":"online","status_checked_at":"2026-05-24T02:00:06.296Z","response_time":57,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","amd-gpu","autoscaling","dynamic-resource-allocation","gpu","gpu-acceleration","gpu-pooling","gpu-scheduling","gpu-usage","gpu-virtualization","inference","karpenter","kubernetes","llm-serving","nvidia","pytorch","rcuda","remote-gpu","vgpu"],"created_at":"2025-03-03T05:16:31.298Z","updated_at":"2026-05-24T04:01:00.918Z","avatar_url":"https://github.com/NexusGPU.png","language":"Go","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\u003ca href=\"javascript:void(0);\" target=\"_blank\" rel=\"noreferrer\"\u003e\u003cimg width=\"100%\" src=\"https://cdn.tensor-fusion.ai/logo-banner.png\" alt=\"Logo\"\u003e\u003c/a\u003e\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n    \u003cbr /\u003e\u003cstrong\u003e\u003ca href=\"https://tensor-fusion.ai\" target=\"_blank\"\u003eTensorFusion.AI\u003c/a\u003e\u003c/strong\u003e\u003cbr/\u003e\u003cb\u003eLess GPUs, More AI Apps.\u003c/b\u003e\n    \u003cbr /\u003e\n    \u003ca href=\"https://tensor-fusion.ai/guide/overview\"\u003e\u003cstrong\u003eExplore the docs »\u003c/strong\u003e\u003c/a\u003e\n    \u003cbr /\u003e\n    \u003ca href=\"https://tensor-fusion.ai/guide/overview\"\u003eView Demo\u003c/a\u003e\n    |\n    \u003ca href=\"https://github.com/NexusGPU/tensor-fusion/issues/new?labels=bug\u0026template=bug-report---.md\"\u003eReport Bug\u003c/a\u003e\n    |\n    \u003ca href=\"https://github.com/NexusGPU/tensor-fusion/issues/new?labels=enhancement\u0026template=feature-request---.md\"\u003eRequest Feature\u003c/a\u003e\n  \u003c/p\u003e\n\n\n[![Contributors][contributors-shield]][contributors-url]\n[![Forks][forks-shield]][forks-url]\n[![Stargazers][stars-shield]][stars-url]\n[![MIT License][license-shield]][license-url]\n[![LinkedIn][linkedin-shield]][linkedin-url]\n[![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/NexusGPU/tensor-fusion)\n\nTensor Fusion is a state-of-the-art **GPU virtualization and pooling solution** designed to optimize GPU cluster utilization to its fullest potential.\n\n## 🌟 Highlights\n\n#### 📐 Fractional Virtual GPU\n#### 🔄 Remote GPU Sharing over Ethernet/InfiniBand\n#### ⚖️ GPU-first Scheduling and Auto-scaling\n#### 📊 GPU Oversubscription and VRAM Expansion\n#### 🛫 GPU Pooling, Monitoring, Live Migration, Model Preloading and more\n\n## 🚀 Quick Start\n\n### Onboard Your Own AI Infra\n\n- [Deploy in Kubernetes with Cloud Console](https://tensor-fusion.ai/guide/getting-started/deployment-k8s)\n- [Deploy in Kubernetes with Helm chart](https://tensor-fusion.ai/guide/recipes/deploy-k8s-local-mode)\n- [Create new cluster in VM/BareMetal](https://tensor-fusion.ai/guide/getting-started/deployment-vm)\n- [Run vGPU in VM Hypervisor](https://tensor-fusion.ai/guide/getting-started/deployment-vm)\n- [Learn Essential Concepts \u0026 Architecture](https://tensor-fusion.ai/guide/getting-started/architecture)\n\n### 💬 Discussion\n\n- Discord channel: [https://discord.gg/2bybv9yQNk](https://discord.gg/2bybv9yQNk)\n- Discuss anything about TensorFusion: [Github Discussions](https://github.com/NexusGPU/tensor-fusion/discussions)\n- Contact us with WeCom for Greater China region: [企业微信](https://work.weixin.qq.com/ca/cawcde42751d9f6a29)\n- Email us: [support@tensor-fusion.com](mailto:support@tensor-fusion.com)\n- Schedule [1:1 meeting with TensorFusion founders](https://tensor-fusion.ai/book-demo)\n\n## 🔮 Features \u0026 Roadmap\n\n### Core GPU Virtualization Features\n\n- [x] Fractional GPU and flexible oversubscription\n- [x] Remote GPU sharing with SOTA GPU-over-IP technology, less than 4% performance loss\n- [x] GPU VRAM expansion and hot/cold tiering\n- [x] Non-NVIDIA GPU/NPU vendor support\n\n### Pooling \u0026 Scheduling \u0026 Management\n\n- [x] GPU/NPU pool management in Kubernetes\n- [x] GPU-first scheduling and allocation, with 1 TFLOPs, 1% Computing, 1 MB precision\n- [x] GPU node auto provisioning/termination, Karpenter integration\n- [x] GPU compaction/bin-packing\n- [x] Take full control of GPU allocation with precision targeting by vendor, model, device index, and more\n- [x] Seamless onboarding experience for PyTorch, TensorFlow, llama.cpp, vLLM, TensorRT, SGLang and all popular AI training/serving frameworks\n- [x] Seamless migration from existing NVIDIA operator and device-plugin stack\n- [x] Centralized Dashboard \u0026 Control Plane\n- [x] GPU-first autoscaling policies, auto set requests/limits/replicas\n- [x] Request multiple vGPUs with group scheduling for large models\n- [x] Support different QoS levels\n- [x] Hardware partitioned mode isolation like NVIDIA Dynamic MIG\n- [x] Support Kubernetes dynamic resource allocation (DRA) API\n\n### Enterprise Features\n\n- [x] GPU live-migration, snapshot and restore GPU context cross cluster\n- [ ] AI model registry and preloading, build your own private MaaS(Model-as-a-Service)\n- [x] Advanced auto-scaling policies, scale to zero, rebalance of hot GPUs\n- [ ] Advanced observability features, detailed metrics \u0026 tracing/profiling of CUDA calls\n- [x] Monetize your GPU cluster by multi-tenancy usage measurement \u0026 billing report\n- [x] Enterprise level high availability and resilience, support topology aware scheduling, GPU node auto failover etc.\n- [x] Enterprise level security, complete on-premise deployment support\n- [ ] Enterprise level compliance, SSO/SAML support, advanced audit, ReBAC control, SOC2 and other compliance reports available\n\n### 🗳️ Platform Support\n\n- [x] Run on Linux Kubernetes clusters\n- [x] Run on Linux VMs or Bare Metal (one-click onboarding to Edge K3S)\n- [x] Run on Windows (Not open sourced, contact us for support)\n\nSee the [open issues](https://github.com/NexusGPU/tensor-fusion/issues) for a full list of proposed features (and known issues).\n\n## 🙏 Contributing\n\nContributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are **greatly appreciated**.\n\nIf you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag \"enhancement\".\nDon't forget to give the project a star! Thanks again!\n\n### Development Setup\n\nThis project patches Kubernetes scheduler components. Before building or running tests, you need to vendor dependencies and apply patches:\n\n```bash\nmake vendor  # Vendor dependencies and apply scheduler patches\n```\n\n\n### How to Contribute\n\n1. Fork the Project\n2. Create your Feature Branch (`git checkout -b feature/AmazingFeature`)\n3. Make your changes and run linter (`make lint-fix`)\n4. Run tests to ensure everything works (`make test`)\n5. Commit your Changes (`git commit -m 'Add some AmazingFeature'`)\n6. Push to the Branch (`git push origin feature/AmazingFeature`)\n7. Open a Pull Request\n\n### Top contributors\n\n\u003ca href=\"https://github.com/NexusGPU/tensor-fusion/graphs/contributors\"\u003e\n  \u003cimg src=\"https://contrib.rocks/image?repo=NexusGPU/tensor-fusion\" alt=\"contrib.rocks image\" /\u003e\n\u003c/a\u003e\n\n## 🔷 License\n\n1. [TensorFusion main repo](https://github.com/NexusGPU/tensor-fusion) is open sourced with [Apache 2.0 License](./LICENSE), which includes **GPU pooling, scheduling, management features**, you can use it for free and customize it as you want.\n2. [vgpu.rs repo](https://github.com/NexusGPU/vgpu.rs) is open sourced with [Apache 2.0 License](./LICENSE), which includes **Fractional GPU** and **vGPU hypervisor features**, you can use it for free and customize it as you want.\n3. **Advanced GPU virtualization and GPU-over-IP sharing features** are also free to use when **GPU total number of your organization is less than 10**, but the implementation is not fully open sourced, please [contact us](mailto:support@tensor-fusion.com) for more details.\n4. Features mentioned in \"**Enterprise Features**\" above are paid, **licensed users can use these features in [TensorFusion Console](https://app.tensor-fusion.ai)**.\n5. For large scale deployment that involves non-free features of #3 and #4, please [contact us](mailto:support@tensor-fusion.com), pricing details are available [here](https://tensor-fusion.ai/pricing)\n\n[![FOSSA Status](https://app.fossa.com/api/projects/git%2Bgithub.com%2FNexusGPU%2Ftensor-fusion.svg?type=large\u0026issueType=license)](https://app.fossa.com/projects/git%2Bgithub.com%2FNexusGPU%2Ftensor-fusion?ref=badge_large\u0026issueType=license)\n\n\u003c!-- MARKDOWN LINKS \u0026 IMAGES --\u003e\n\u003c!-- https://www.markdownguide.org/basic-syntax/#reference-style-links --\u003e\n[contributors-shield]: https://img.shields.io/github/contributors/NexusGPU/tensor-fusion.svg?style=for-the-badge\n[contributors-url]: https://github.com/NexusGPU/tensor-fusion/graphs/contributors\n[forks-shield]: https://img.shields.io/github/forks/NexusGPU/tensor-fusion.svg?style=for-the-badge\n[forks-url]: https://github.com/NexusGPU/tensor-fusion/network/members\n[stars-shield]: https://img.shields.io/github/stars/NexusGPU/tensor-fusion.svg?style=for-the-badge\n[stars-url]: https://github.com/NexusGPU/tensor-fusion/stargazers\n[issues-shield]: https://img.shields.io/github/issues/NexusGPU/tensor-fusion.svg?style=for-the-badge\n[issues-url]: https://github.com/NexusGPU/tensor-fusion/issues\n[license-shield]: https://img.shields.io/github/license/NexusGPU/tensor-fusion.svg?style=for-the-badge\n[license-url]: https://github.com/NexusGPU/tensor-fusion/blob/master/LICENSE\n[linkedin-shield]: https://img.shields.io/badge/-LinkedIn-black.svg?style=for-the-badge\u0026logo=linkedin\u0026colorB=555\n[linkedin-url]: https://www.linkedin.com/company/tensor-fusion/about\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnexusgpu%2Ftensor-fusion","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnexusgpu%2Ftensor-fusion","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnexusgpu%2Ftensor-fusion/lists"}