[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-9920":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":23,"hasPages":25,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":46,"readmeContent":47,"aiSummary":48,"trendingCount":16,"starSnapshotCount":16,"syncStatus":49,"lastSyncTime":50,"discoverSource":51},9920,"flyte","flyteorg\u002Fflyte","flyteorg","Dynamic, resilient AI orchestration. Coordinate data, models, and compute as you build AI workflows.","https:\u002F\u002Fflyte.org",null,"Go",7079,829,249,93,0,1,16,60,5,80.76,"Apache License 2.0",false,"main",true,[27,28,29,30,31,32,33,5,34,35,36,37,38,39,40,41,42,43,44,45],"agentic","ai-agents","ai-development-tools","data-analysis","data-science","declarative","fine-tuning","golang","grpc","hacktoberfest","kubernetes","llm","machine-learning","mlops","orchestration-engine","production","python","scale","workflow","2026-06-12 04:00:47","> [!IMPORTANT]\n> ## Flyte 2 Devbox is now available!\n>\n> Check out the guide [here](https:\u002F\u002Fwww.union.ai\u002Fdocs\u002Fv2\u002Fflyte\u002Fuser-guide\u002Frun-modes\u002Frunning-devbox\u002F) to get started.\n>\n> Looking for Flyte 1? Go to the [master](https:\u002F\u002Fgithub.com\u002Fflyteorg\u002Fflyte\u002Ftree\u002Fmaster) branch, where Flyte 1 is now maintained.\n\n---\n\n# Flyte 2\n\n**Reliably orchestrate ML pipelines, models, and agents at scale — in pure Python.**\n\n[![Version](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fflyte?label=version&color=blue)](https:\u002F\u002Fpypi.org\u002Fproject\u002Fflyte\u002F)\n[![Python](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fpyversions\u002Fflyte?color=brightgreen)](https:\u002F\u002Fpypi.org\u002Fproject\u002Fflyte\u002F)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-Apache%202.0-orange)](LICENSE)\n[![Try in Browser](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTry%20in%20Browser-Live%20Demo-7652a2)](https:\u002F\u002Fflyte2intro.apps.demo.hosted.unionai.cloud\u002F)\n[![Docs](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDocs-flyte-blue)](https:\u002F\u002Fwww.union.ai\u002Fdocs\u002Fv2\u002Fflyte\u002Fuser-guide\u002Frunning-locally\u002F)\n[![SDK Reference](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSDK%20Reference-API-brightgreen)](https:\u002F\u002Fwww.union.ai\u002Fdocs\u002Fv2\u002Fbyoc\u002Fapi-reference\u002Fflyte-sdk\u002F)\n[![CLI Reference](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCLI%20Reference-API-brightgreen)](https:\u002F\u002Fwww.union.ai\u002Fdocs\u002Fv2\u002Fbyoc\u002Fapi-reference\u002Fflyte-cli\u002F)\n\nFlyte is a Graduated project of the [LF AI & Data Foundation](https:\u002F\u002Flfaidata.foundation\u002Fprojects\u002Fflyte\u002F).\n\n\u003Ca href=\"https:\u002F\u002Flfaidata.foundation\u002Fprojects\u002Fflyte\u002F\">\n  \u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fflyteorg\u002Fstatic-resources\u002Fmain\u002Fflyte\u002Freadme\u002Fflyte_and_lf.png\" alt=\"Flyte and LF AI & Data Logo\" width=\"250\">\n\u003C\u002Fa>\n\n## Install\n\n```bash\nuv pip install flyte\n```\n\nFor the full SDK and development tools, see the [flyte-sdk](https:\u002F\u002Fgithub.com\u002Fflyteorg\u002Fflyte-sdk) repository.\n\n## Example\n\n```python\nimport asyncio\nimport flyte\n\nenv = flyte.TaskEnvironment(\n    name=\"hello_world\",\n    image=flyte.Image.from_debian_base(python_version=(3, 12)),\n)\n\n@env.task\ndef calculate(x: int) -> int:\n    return x * 2 + 5\n\n@env.task\nasync def main(numbers: list[int]) -> float:\n    results = await asyncio.gather(*[\n        calculate.aio(num) for num in numbers\n    ])\n    return sum(results) \u002F len(results)\n\nif __name__ == \"__main__\":\n    flyte.init()\n    run = flyte.run(main, numbers=list(range(10)))\n    print(f\"Result: {run.result}\")\n```\n\n\u003Ctable>\n\u003Ctr>\u003Ctd>\u003Cb>Python\u003C\u002Fb>\u003C\u002Ftd>\u003Ctd>\u003Cb>Flyte CLI\u003C\u002Fb>\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\n\n```bash\npython hello.py\n```\n\n\u003C\u002Ftd>\n\u003Ctd>\n\n```bash\nflyte run hello.py main --numbers '[1,2,3]'\n```\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftable>\n\n## Serve a Model\n\n```python\n# serving.py\nfrom fastapi import FastAPI\nimport flyte\nfrom flyte.app.extras import FastAPIAppEnvironment\n\napp = FastAPI()\nenv = FastAPIAppEnvironment(\n    name=\"my-model\",\n    app=app,\n    image=flyte.Image.from_debian_base(python_version=(3, 12)).with_pip_packages(\n        \"fastapi\", \"uvicorn\"\n    ),\n)\n\n@app.get(\"\u002Fpredict\")\nasync def predict(x: float) -> dict:\n    return {\"result\": x * 2 + 5}\n\nif __name__ == \"__main__\":\n    flyte.init_from_config()\n    flyte.serve(env)\n```\n\n\u003Ctable>\n\u003Ctr>\u003Ctd>\u003Cb>Python\u003C\u002Fb>\u003C\u002Ftd>\u003Ctd>\u003Cb>Flyte CLI\u003C\u002Fb>\u003C\u002Ftd>\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\n\n```bash\npython serving.py\n```\n\n\u003C\u002Ftd>\n\u003Ctd>\n\n```bash\nflyte serve serving.py env\n```\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftable>\n\n## Local Development Experience\n\nInstall the TUI for a rich local development experience:\n\n```bash\nuv pip install flyte[tui]\n```\n\n[![Watch the local development experience](https:\u002F\u002Fimg.youtube.com\u002Fvi\u002Flsfy-7DbbRM\u002Fmaxresdefault.jpg)](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=lsfy-7DbbRM)\n\n**[Try the hosted demo in your browser](https:\u002F\u002Fflyte2intro.apps.demo.hosted.unionai.cloud\u002F)** — no installation required.\n\n## Open Source Backend\n\nThe open source backend for Flyte 2 is **coming soon**. This repository will contain the Kubernetes-native backend infrastructure for deploying Flyte 2 as a distributed, multi-node service. See the [Backend README](docs\u002FBACKEND_README.md) for the current state of the backend, protocol buffer definitions, and contribution guide.\n\nIf you need an enterprise-ready, production-grade backend for Flyte 2 today, it is available on [Union.ai](https:\u002F\u002Fwww.union.ai\u002Ftry-flyte-2).\n\n## Learn More\n\n- **[Live Demo](https:\u002F\u002Fflyte2intro.apps.demo.hosted.unionai.cloud\u002F)** — Try Flyte 2 in your browser\n- **[Documentation](https:\u002F\u002Fwww.union.ai\u002Fdocs\u002Fv2\u002Fflyte\u002Fuser-guide\u002Frunning-locally\u002F)** — Get started running locally\n- **[SDK Reference](https:\u002F\u002Fwww.union.ai\u002Fdocs\u002Fv2\u002Fbyoc\u002Fapi-reference\u002Fflyte-sdk\u002F)** — API reference docs\n- **[CLI Reference](https:\u002F\u002Fwww.union.ai\u002Fdocs\u002Fv2\u002Fbyoc\u002Fapi-reference\u002Fflyte-cli\u002F)** — CLI docs\n- **[flyte-sdk](https:\u002F\u002Fgithub.com\u002Fflyteorg\u002Fflyte-sdk)** — The Flyte 2 Python SDK repository\n- **[Join the Flyte 2 Production Preview](https:\u002F\u002Fwww.union.ai\u002Ftry-flyte-2)** — Get early access\n- **[Slack](https:\u002F\u002Fslack.flyte.org\u002F)** | **[GitHub Discussions](https:\u002F\u002Fgithub.com\u002Fflyteorg\u002Fflyte\u002Fdiscussions)** | **[Issues](https:\u002F\u002Fgithub.com\u002Fflyteorg\u002Fflyte\u002Fissues)**\n\n## Contributing\n\nWe welcome contributions! See the [Backend README](docs\u002FBACKEND_README.md) for backend development, or join us on [slack.flyte.org](https:\u002F\u002Fslack.flyte.org).\n\n## Sponsors\n\nCI container image builds for this repository are sponsored by [Depot](https:\u002F\u002Fdepot.dev) — fast, native multi-arch Docker builds with persistent layer caching.\n\n[![Built with Depot](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBuilt%20with-Depot-7c3aed?style=for-the-badge&logo=docker&logoColor=white)](https:\u002F\u002Fdepot.dev)\n\n## License\n\nApache 2.0 — see [LICENSE](LICENSE).\n","Flyte 是一个用于构建和协调 AI 工作流的动态且可靠的编排引擎，支持数据、模型和计算资源的协同。其核心功能包括使用纯 Python 编写复杂的机器学习管道，并通过声明式 API 来定义任务和工作流。技术特点上，Flyte 采用了 Go 语言编写，并支持 gRPC 和 Kubernetes 等现代云原生技术栈，确保了高可用性和可扩展性。它特别适合需要大规模部署和管理机器学习模型的企业级应用场景，以及追求高效开发和迭代的数据科学家与工程师团队。",2,"2026-06-11 03:25:25","top_topic"]