[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-microsoft--promptflow":3,"tool-microsoft--promptflow":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",142651,2,"2026-04-06T23:34:12",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107888,"2026-04-06T11:32:50",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":32,"last_commit_at":50,"category_tags":51,"status":17},4721,"markitdown","microsoft\u002Fmarkitdown","MarkItDown 是一款由微软 AutoGen 团队打造的轻量级 Python 工具，专为将各类文件高效转换为 Markdown 格式而设计。它支持 PDF、Word、Excel、PPT、图片（含 OCR）、音频（含语音转录）、HTML 乃至 YouTube 链接等多种格式的解析，能够精准提取文档中的标题、列表、表格和链接等关键结构信息。\n\n在人工智能应用日益普及的今天，大语言模型（LLM）虽擅长处理文本，却难以直接读取复杂的二进制办公文档。MarkItDown 恰好解决了这一痛点，它将非结构化或半结构化的文件转化为模型“原生理解”且 Token 效率极高的 Markdown 格式，成为连接本地文件与 AI 分析 pipeline 的理想桥梁。此外，它还提供了 MCP（模型上下文协议）服务器，可无缝集成到 Claude Desktop 等 LLM 应用中。\n\n这款工具特别适合开发者、数据科学家及 AI 研究人员使用，尤其是那些需要构建文档检索增强生成（RAG）系统、进行批量文本分析或希望让 AI 助手直接“阅读”本地文件的用户。虽然生成的内容也具备一定可读性，但其核心优势在于为机器",93400,"2026-04-06T19:52:38",[52,14],"插件",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":10,"last_commit_at":59,"category_tags":60,"status":17},4487,"LLMs-from-scratch","rasbt\u002FLLMs-from-scratch","LLMs-from-scratch 是一个基于 PyTorch 的开源教育项目，旨在引导用户从零开始一步步构建一个类似 ChatGPT 的大型语言模型（LLM）。它不仅是同名技术著作的官方代码库，更提供了一套完整的实践方案，涵盖模型开发、预训练及微调的全过程。\n\n该项目主要解决了大模型领域“黑盒化”的学习痛点。许多开发者虽能调用现成模型，却难以深入理解其内部架构与训练机制。通过亲手编写每一行核心代码，用户能够透彻掌握 Transformer 架构、注意力机制等关键原理，从而真正理解大模型是如何“思考”的。此外，项目还包含了加载大型预训练权重进行微调的代码，帮助用户将理论知识延伸至实际应用。\n\nLLMs-from-scratch 特别适合希望深入底层原理的 AI 开发者、研究人员以及计算机专业的学生。对于不满足于仅使用 API，而是渴望探究模型构建细节的技术人员而言，这是极佳的学习资源。其独特的技术亮点在于“循序渐进”的教学设计：将复杂的系统工程拆解为清晰的步骤，配合详细的图表与示例，让构建一个虽小但功能完备的大模型变得触手可及。无论你是想夯实理论基础，还是为未来研发更大规模的模型做准备",90106,"2026-04-06T11:19:32",[35,15,13,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":76,"owner_email":77,"owner_twitter":78,"owner_website":79,"owner_url":80,"languages":81,"stars":117,"forks":118,"last_commit_at":119,"license":120,"difficulty_score":32,"env_os":121,"env_gpu":121,"env_ram":121,"env_deps":122,"category_tags":127,"github_topics":128,"view_count":32,"oss_zip_url":76,"oss_zip_packed_at":76,"status":17,"created_at":137,"updated_at":138,"faqs":139,"releases":168},4959,"microsoft\u002Fpromptflow","promptflow","Build high-quality LLM apps - from prototyping, testing to production deployment and monitoring.","Prompt flow 是一套专为大语言模型（LLM）应用打造的全流程开发工具，旨在帮助用户轻松完成从创意原型、测试评估到生产部署及监控的完整闭环。它主要解决了大模型应用在开发过程中流程分散、调试困难以及质量难以量化等痛点，让提示词工程变得更加系统化且高效。\n\n无论是希望快速验证想法的 AI 研究人员，还是致力于构建高质量生产级应用的开发者，都能通过 Prompt flow 获益。其核心亮点在于支持将大模型、提示词、Python 代码及其他工具灵活编排为可执行的“流（Flow）”，并提供直观的追踪功能，让用户能轻松调试与大模型的交互细节。此外，它还内置了强大的评估机制，支持利用大规模数据集对应用性能进行量化测试，并能无缝集成至 CI\u002FCD 系统以保障交付质量。配合丰富的命令行工具和 Visual Studio Code 扩展，Prompt flow 让构建可靠的大模型应用变得简单而专业。","# Prompt flow\n\n[![Python package](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fpromptflow)](https:\u002F\u002Fpypi.org\u002Fproject\u002Fpromptflow\u002F)\n[![Python](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fpyversions\u002Fpromptflow.svg?maxAge=2592000)](https:\u002F\u002Fpypi.python.org\u002Fpypi\u002Fpromptflow\u002F)\n[![PyPI - Downloads](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdm\u002Fpromptflow)](https:\u002F\u002Fpypi.org\u002Fproject\u002Fpromptflow\u002F)\n[![CLI](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCLI-reference-blue)](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Freference\u002Fpf-command-reference.html)\n[![vsc extension](https:\u002F\u002Fimg.shields.io\u002Fvisual-studio-marketplace\u002Fi\u002Fprompt-flow.prompt-flow?logo=Visual%20Studio&label=Extension%20)](https:\u002F\u002Fmarketplace.visualstudio.com\u002Fitems?itemName=prompt-flow.prompt-flow)\n\n[![Doc](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDoc-online-green)](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Findex.html)\n[![Issue](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002Fmicrosoft\u002Fpromptflow)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002Fnew\u002Fchoose)\n[![Discussions](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fdiscussions\u002Fmicrosoft\u002Fpromptflow)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002Fnew\u002Fchoose)\n[![CONTRIBUTING](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FContributing-8A2BE2)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Fmain\u002FCONTRIBUTING.md)\n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002Fmicrosoft\u002Fpromptflow)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Fmain\u002FLICENSE)\n\n> Welcome to join us to make prompt flow better by\n> participating [discussions](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fdiscussions),\n> opening [issues](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002Fnew\u002Fchoose),\n> submitting [PRs](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fpulls).\n\n**Prompt flow** is a suite of development tools designed to streamline the end-to-end development cycle of LLM-based AI applications, from ideation, prototyping, testing, evaluation to production deployment and monitoring. It makes prompt engineering much easier and enables you to build LLM apps with production quality.\n\nWith prompt flow, you will be able to:\n\n- **Create and iteratively develop flow**\n    - Create executable [flows](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Fconcepts\u002Fconcept-flows.html) that link LLMs, prompts, Python code and other [tools](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Fconcepts\u002Fconcept-tools.html) together.\n    - Debug and iterate your flows, especially [tracing interaction with LLMs](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Fhow-to-guides\u002Ftracing\u002Findex.html) with ease.\n- **Evaluate flow quality and performance**\n    - Evaluate your flow's quality and performance with larger datasets.\n    - Integrate the testing and evaluation into your CI\u002FCD system to ensure quality of your flow.\n- **Streamlined development cycle for production**\n    - Deploy your flow to the serving platform you choose or integrate into your app's code base easily.\n    - (Optional but highly recommended) Collaborate with your team by leveraging the cloud version of [Prompt flow in Azure AI](https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fazure\u002Fmachine-learning\u002Fprompt-flow\u002Foverview-what-is-prompt-flow?view=azureml-api-2).\n\n------\n\n## Installation\n\nTo get started quickly, you can use a pre-built development environment. **Click the button below** to open the repo in GitHub Codespaces, and then continue the readme!\n\n[![Open in GitHub Codespaces](https:\u002F\u002Fgithub.com\u002Fcodespaces\u002Fbadge.svg)](https:\u002F\u002Fcodespaces.new\u002Fmicrosoft\u002Fpromptflow?quickstart=1)\n\nIf you want to get started in your local environment, first install the packages:\n\nEnsure you have a python environment, `python>=3.9, \u003C=3.11` is recommended.\n\n```sh\npip install promptflow promptflow-tools\n```\n\n## Quick Start ⚡\n\n**Create a chatbot with prompt flow**\n\nRun the command to initiate a prompt flow from a chat template, it creates folder named `my_chatbot` and generates required files within it:\n\n```sh\npf flow init --flow .\u002Fmy_chatbot --type chat\n```\n\n**Setup a connection for your API key**\n\nFor OpenAI key, establish a connection by running the command, using the `openai.yaml` file in the `my_chatbot` folder, which stores your OpenAI key (override keys and name with --set to avoid yaml file changes):\n\n```sh\npf connection create --file .\u002Fmy_chatbot\u002Fopenai.yaml --set api_key=\u003Cyour_api_key> --name open_ai_connection\n```\n\nFor Azure OpenAI key, establish the connection by running the command, using the `azure_openai.yaml` file:\n\n```sh\npf connection create --file .\u002Fmy_chatbot\u002Fazure_openai.yaml --set api_key=\u003Cyour_api_key> api_base=\u003Cyour_api_base> --name open_ai_connection\n```\n\n**Chat with your flow**\n\nIn the `my_chatbot` folder, there's a `flow.dag.yaml` file that outlines the flow, including inputs\u002Foutputs, nodes,  connection, and the LLM model, etc\n\n> Note that in the `chat` node, we're using a connection named `open_ai_connection` (specified in `connection` field) and the `gpt-35-turbo` model (specified in `deployment_name` field). The deployment_name filed is to specify the OpenAI model, or the Azure OpenAI deployment resource.\n\nInteract with your chatbot by running: (press `Ctrl + C` to end the session)\n\n```sh\npf flow test --flow .\u002Fmy_chatbot --interactive\n```\n\n**Core value: ensuring \"High Quality” from prototype to production**\n\nExplore our [**15-minute tutorial**](examples\u002Ftutorials\u002Fflow-fine-tuning-evaluation\u002Fpromptflow-quality-improvement.md) that guides you through prompt tuning ➡ batch testing ➡ evaluation, all designed to ensure high quality ready for production.\n\nNext Step! Continue with the **Tutorial**  👇 section to delve deeper into prompt flow.\n\n## Tutorial 🏃‍♂️\n\nPrompt flow is a tool designed to **build high quality LLM apps**, the development process in prompt flow follows these steps: develop a flow, improve the flow quality, deploy the flow to production.\n\n### Develop your own LLM apps\n\n#### VS Code Extension\n\nWe also offer a VS Code extension (a flow designer) for an interactive flow development experience with UI.\n\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fmicrosoft_promptflow_readme_b007eb73eb56.png\" alt=\"vsc\" width=\"1000\"\u002F>\n\nYou can install it from the \u003Ca href=\"https:\u002F\u002Fmarketplace.visualstudio.com\u002Fitems?itemName=prompt-flow.prompt-flow\">visualstudio marketplace\u003C\u002Fa>.\n\n#### Deep delve into flow development\n\n[Getting started with prompt flow](.\u002Fdocs\u002Fhow-to-guides\u002Fquick-start.md): A step by step guidance to invoke your first flow run.\n\n### Learn from use cases\n\n[Tutorial: Chat with PDF](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Fmain\u002Fexamples\u002Ftutorials\u002Fe2e-development\u002Fchat-with-pdf.md): An end-to-end tutorial on how to build a high quality chat application with prompt flow, including flow development and evaluation with metrics.\n> More examples can be found [here](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Ftutorials\u002Findex.html#samples). We welcome contributions of new use cases!\n\n### Setup for contributors\n\nIf you're interested in contributing, please start with our dev setup guide: [dev_setup.md](.\u002Fdocs\u002Fdev\u002Fdev_setup.md).\n\nNext Step! Continue with the **Contributing**  👇 section to contribute to prompt flow.\n\n## Contributing\n\nThis project welcomes contributions and suggestions.  Most contributions require you to agree to a\nContributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us\nthe rights to use your contribution. For details, visit https:\u002F\u002Fcla.opensource.microsoft.com.\n\nWhen you submit a pull request, a CLA bot will automatically determine whether you need to provide\na CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions\nprovided by the bot. You will only need to do this once across all repos using our CLA.\n\nThis project has adopted the [Microsoft Open Source Code of Conduct](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002F).\nFor more information see the [Code of Conduct FAQ](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002Ffaq\u002F) or\ncontact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional questions or comments.\n\n## Trademarks\n\nThis project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft\ntrademarks or logos is subject to and must follow\n[Microsoft's Trademark & Brand Guidelines](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Flegal\u002Fintellectualproperty\u002Ftrademarks\u002Fusage\u002Fgeneral).\nUse of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship.\nAny use of third-party trademarks or logos are subject to those third-party's policies.\n\n## Code of Conduct\n\nThis project has adopted the\n[Microsoft Open Source Code of Conduct](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002F).\nFor more information see the\n[Code of Conduct FAQ](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002Ffaq\u002F)\nor contact [opencode@microsoft.com](mailto:opencode@microsoft.com)\nwith any additional questions or comments.\n\n## Data Collection\n\nThe software may collect information about you and your use of the software and\nsend it to Microsoft if configured to enable telemetry.\nMicrosoft may use this information to provide services and improve our products and services.\nYou may turn on the telemetry as described in the repository.\nThere are also some features in the software that may enable you and Microsoft\nto collect data from users of your applications. If you use these features, you\nmust comply with applicable law, including providing appropriate notices to\nusers of your applications together with a copy of Microsoft's privacy\nstatement. Our privacy statement is located at\nhttps:\u002F\u002Fgo.microsoft.com\u002Ffwlink\u002F?LinkID=824704. You can learn more about data\ncollection and use in the help documentation and our privacy statement. Your\nuse of the software operates as your consent to these practices.\n\n### Telemetry Configuration\n\nTelemetry collection is on by default.\n\nTo opt out, please run `pf config set telemetry.enabled=false` to turn it off.\n\n## License\n\nCopyright (c) Microsoft Corporation. All rights reserved.\n\nLicensed under the [MIT](LICENSE) license.\n","# 提示流程\n\n[![Python 包](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fpromptflow)](https:\u002F\u002Fpypi.org\u002Fproject\u002Fpromptflow\u002F)\n[![Python](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fpyversions\u002Fpromptflow.svg?maxAge=2592000)](https:\u002F\u002Fpypi.python.org\u002Fpypi\u002Fpromptflow\u002F)\n[![PyPI - 下载量](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdm\u002Fpromptflow)](https:\u002F\u002Fpypi.org\u002Fproject\u002Fpromptflow\u002F)\n[![CLI](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCLI-参考-blue)](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Freference\u002Fpf-command-reference.html)\n[![VS Code 扩展](https:\u002F\u002Fimg.shields.io\u002Fvisual-studio-marketplace\u002Fi\u002Fprompt-flow.prompt-flow?logo=Visual%20Studio&label=Extension%20)](https:\u002F\u002Fmarketplace.visualstudio.com\u002Fitems?itemName=prompt-flow.prompt-flow)\n\n[![文档](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDoc-在线-green)](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Findex.html)\n[![问题](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002Fmicrosoft\u002Fpromptflow)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002Fnew\u002Fchoose)\n[![讨论](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fdiscussions\u002Fmicrosoft\u002Fpromptflow)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002Fnew\u002Fchoose)\n[![贡献指南](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FContributing-8A2BE2)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Fmain\u002FCONTRIBUTING.md)\n[![许可证：MIT](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002Fmicrosoft\u002Fpromptflow)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Fmain\u002FLICENSE)\n\n> 欢迎加入我们，通过参与 [讨论](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fdiscussions)、提交 [问题](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002Fnew\u002Fchoose) 和 [拉取请求](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fpulls) 来共同让提示流程更加完善。\n\n**提示流程** 是一套开发工具，旨在简化基于 LLM 的 AI 应用程序的端到端开发周期，从构思、原型设计、测试、评估到生产部署和监控。它使提示工程变得更加容易，并帮助您构建具有生产质量的 LLM 应用程序。\n\n借助提示流程，您可以：\n\n- **创建并迭代开发流程**\n    - 创建可执行的 [流程](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Fconcepts\u002Fconcept-flows.html)，将 LLM、提示、Python 代码和其他 [工具](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Fconcepts\u002Fconcept-tools.html) 连接在一起。\n    - 轻松调试和迭代您的流程，尤其是 [跟踪与 LLM 的交互](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Fhow-to-guides\u002Ftracing\u002Findex.html)。\n- **评估流程质量和性能**\n    - 使用更大的数据集评估流程的质量和性能。\n    - 将测试和评估集成到您的 CI\u002FCD 系统中，以确保流程的质量。\n- **为生产环境优化开发流程**\n    - 将您的流程部署到您选择的服务平台，或轻松集成到您的应用程序代码库中。\n    - （可选但强烈推荐）通过利用 Azure AI 中的云版本 [Prompt Flow](https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fazure\u002Fmachine-learning\u002Fprompt-flow\u002Foverview-what-is-prompt-flow?view=azureml-api-2) 与团队协作。\n\n------\n\n## 安装\n\n要快速入门，您可以使用预构建的开发环境。**点击下方按钮** 在 GitHub Codespaces 中打开仓库，然后继续阅读自述文件！\n\n[![在 GitHub Codespaces 中打开](https:\u002F\u002Fgithub.com\u002Fcodespaces\u002Fbadge.svg)](https:\u002F\u002Fcodespaces.new\u002Fmicrosoft\u002Fpromptflow?quickstart=1)\n\n如果您想在本地环境中开始使用，首先安装以下包：\n\n请确保您有一个 Python 环境，建议使用 `python>=3.9, \u003C=3.11`。\n\n```sh\npip install promptflow promptflow-tools\n```\n\n## 快速入门 ⚡\n\n**使用提示流程创建聊天机器人**\n\n运行命令从聊天模板初始化一个提示流程，它会创建名为 `my_chatbot` 的文件夹，并在其中生成所需的文件：\n\n```sh\npf flow init --flow .\u002Fmy_chatbot --type chat\n```\n\n**为您的 API 密钥设置连接**\n\n对于 OpenAI 密钥，请运行以下命令建立连接，使用 `my_chatbot` 文件夹中的 `openai.yaml` 文件存储您的 OpenAI 密钥（使用 `--set` 参数覆盖密钥和名称，以避免修改 YAML 文件）：\n\n```sh\npf connection create --file .\u002Fmy_chatbot\u002Fopenai.yaml --set api_key=\u003Cyour_api_key> --name open_ai_connection\n```\n\n对于 Azure OpenAI 密钥，请运行以下命令建立连接，使用 `azure_openai.yaml` 文件：\n\n```sh\npf connection create --file .\u002Fmy_chatbot\u002Fazure_openai.yaml --set api_key=\u003Cyour_api_key> api_base=\u003Cyour_api_base> --name open_ai_connection\n```\n\n**与您的流程对话**\n\n在 `my_chatbot` 文件夹中，有一个 `flow.dag.yaml` 文件，概述了流程，包括输入\u002F输出、节点、连接以及 LLM 模型等。\n\n> 请注意，在 `chat` 节点中，我们使用名为 `open_ai_connection` 的连接（在 `connection` 字段中指定）和 `gpt-35-turbo` 模型（在 `deployment_name` 字段中指定）。`deployment_name` 字段用于指定 OpenAI 模型，或 Azure OpenAI 部署资源。\n\n通过运行以下命令与您的聊天机器人互动：（按 `Ctrl + C` 结束会话）\n\n```sh\npf flow test --flow .\u002Fmy_chatbot --interactive\n```\n\n**核心价值：从原型到生产确保“高质量”**\n\n探索我们的 [**15 分钟教程**](examples\u002Ftutorials\u002Fflow-fine-tuning-evaluation\u002Fpromptflow-quality-improvement.md)，该教程将指导您完成提示微调 ➡ 批量测试 ➡ 评估的过程，所有这些步骤都旨在确保高质量，为生产做好准备。\n\n下一步！继续阅读下方的 **教程** 部分，深入了解提示流程。\n\n## 教程 🏃‍♂️\n\n提示流程是一种用于 **构建高质量 LLM 应用程序** 的工具，其开发流程遵循以下步骤：开发流程、提升流程质量、将流程部署到生产环境。\n\n### 开发您自己的 LLM 应用程序\n\n#### VS Code 扩展\n\n我们还提供一个 VS Code 扩展（流程设计器），让您可以通过 UI 体验交互式的流程开发。\n\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fmicrosoft_promptflow_readme_b007eb73eb56.png\" alt=\"vsc\" width=\"1000\"\u002F>\n\n您可以从 \u003Ca href=\"https:\u002F\u002Fmarketplace.visualstudio.com\u002Fitems?itemName=prompt-flow.prompt-flow\">Visual Studio 市场\u003C\u002Fa> 安装它。\n\n#### 深入了解流程开发\n\n[提示流程入门](.\u002Fdocs\u002Fhow-to-guides\u002Fquick-start.md)：逐步指导您运行第一个流程。\n\n### 从用例中学习\n\n[教程：与 PDF 对话](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Fmain\u002Fexamples\u002Ftutorials\u002Fe2e-development\u002Fchat-with-pdf.md)：一个端到端教程，介绍如何使用提示流程构建高质量的聊天应用程序，包括流程开发和基于指标的评估。\n> 更多示例可以在这里找到 [这里](https:\u002F\u002Fmicrosoft.github.io\u002Fpromptflow\u002Ftutorials\u002Findex.html#samples)。我们欢迎新的用例贡献！\n\n### 贡献者设置\n\n如果您有兴趣贡献代码，请从我们的开发设置指南开始：[dev_setup.md](.\u002Fdocs\u002Fdev\u002Fdev_setup.md)。\n\n下一步！继续阅读下方的 **贡献** 部分，为提示流程贡献力量。\n\n## 贡献\n\n本项目欢迎贡献和建议。大多数贡献都需要您签署贡献者许可协议（CLA），声明您有权且确实将您的贡献使用权授予我们。有关详细信息，请访问 https:\u002F\u002Fcla.opensource.microsoft.com。\n\n当您提交拉取请求时，CLA 机器人会自动判断您是否需要提供 CLA，并相应地标记 PR（例如状态检查、评论）。请按照机器人提供的指示操作。对于使用我们 CLA 的所有仓库，您只需执行此操作一次。\n\n本项目已采用 [微软开源行为准则](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002F)。更多信息请参阅 [行为准则常见问题解答](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002Ffaq\u002F) 或发送电子邮件至 [opencode@microsoft.com](mailto:opencode@microsoft.com) 咨询更多问题或意见。\n\n## 商标\n\n本项目可能包含项目、产品或服务的商标或标识。未经授权使用微软商标或标识须遵守并遵循 [微软商标与品牌指南](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Flegal\u002Fintellectualproperty\u002Ftrademarks\u002Fusage\u002Fgeneral)。在本项目的修改版本中使用微软商标或标识不得造成混淆或暗示微软的赞助关系。任何第三方商标或标识的使用均应遵守该第三方的相关政策。\n\n## 行为准则\n\n本项目已采用\n[微软开源行为准则](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002F)。\n更多信息请参阅\n[行为准则常见问题解答](https:\u002F\u002Fopensource.microsoft.com\u002Fcodeofconduct\u002Ffaq\u002F)\n或发送电子邮件至 [opencode@microsoft.com](mailto:opencode@microsoft.com)\n咨询更多问题或意见。\n\n## 数据收集\n\n如果配置为启用遥测功能，本软件可能会收集有关您及您使用该软件的信息，并将其发送给微软。微软可能会利用这些信息来提供服务并改进其产品和服务。您可以在仓库中找到启用遥测功能的方法。此外，本软件中还有一些功能可以帮助您和微软收集您应用程序用户的数据。如果您使用这些功能，必须遵守适用法律，包括向您的应用程序用户发出适当的通知，并附上微软的隐私声明。我们的隐私声明位于 https:\u002F\u002Fgo.microsoft.com\u002Ffwlink\u002F?LinkID=824704。您可以通过帮助文档和我们的隐私声明了解更多关于数据收集和使用的信息。您使用本软件即表示您同意这些做法。\n\n### 遥测配置\n\n默认情况下，遥测收集功能已开启。\n\n如需退出，请运行 `pf config set telemetry.enabled=false` 来关闭它。\n\n## 许可证\n\n版权所有 © 微软公司。保留所有权利。\n\n根据 [MIT 许可证](LICENSE) 授权。","# Prompt Flow 快速上手指南\n\nPrompt Flow 是一套旨在简化基于大语言模型（LLM）的 AI 应用端到端开发周期的工具套件。它涵盖了从构思、原型设计、测试、评估到生产部署和监控的全过程，帮助开发者轻松进行提示工程并构建高质量的生产级 LLM 应用。\n\n## 环境准备\n\n在开始之前，请确保您的开发环境满足以下要求：\n\n*   **操作系统**：Windows, macOS 或 Linux\n*   **Python 版本**：推荐 `3.9` 至 `3.11` (必须 >=3.9 且 \u003C=3.11)\n*   **前置依赖**：已安装 `pip` 包管理工具\n\n> **提示**：如果您希望快速体验而无需配置本地环境，可以直接点击 [GitHub Codespaces](https:\u002F\u002Fcodespaces.new\u002Fmicrosoft\u002Fpromptflow?quickstart=1) 在浏览器中打开预构建的开发环境。\n\n## 安装步骤\n\n使用 pip 安装核心库及常用工具包。国内用户若遇到下载缓慢，可添加国内镜像源（如清华源或阿里源）。\n\n**标准安装命令：**\n```sh\npip install promptflow promptflow-tools\n```\n\n**使用国内镜像源加速安装（推荐）：**\n```sh\npip install promptflow promptflow-tools -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n```\n\n## 基本使用\n\n以下是最简单的快速启动流程，我们将创建一个聊天机器人应用，配置 API 连接并进行交互测试。\n\n### 1. 初始化聊天流\n运行以下命令创建一个名为 `my_chatbot` 的文件夹，并生成基于聊天模板的必要文件：\n\n```sh\npf flow init --flow .\u002Fmy_chatbot --type chat\n```\n\n### 2. 配置 API 连接\n您需要将 OpenAI 或 Azure OpenAI 的密钥配置为连接对象。\n\n**对于 OpenAI 用户：**\n使用生成的 `openai.yaml` 文件创建连接（请将 `\u003Cyour_api_key>` 替换为您的实际密钥）：\n```sh\npf connection create --file .\u002Fmy_chatbot\u002Fopenai.yaml --set api_key=\u003Cyour_api_key> --name open_ai_connection\n```\n\n**对于 Azure OpenAI 用户：**\n使用 `azure_openai.yaml` 文件创建连接（需替换密钥和基础 URL）：\n```sh\npf connection create --file .\u002Fmy_chatbot\u002Fazure_openai.yaml --set api_key=\u003Cyour_api_key> api_base=\u003Cyour_api_base> --name open_ai_connection\n```\n\n> **注意**：上述命令中的 `--name` 参数定义了连接名称，需与后续流程配置中的名称保持一致。默认模板中使用的模型部署名为 `gpt-35-turbo`，请根据您的实际资源进行调整。\n\n### 3. 交互式测试\n进入 `my_chatbot` 目录或直接指定路径，运行以下命令启动交互式聊天会话：\n\n```sh\npf flow test --flow .\u002Fmy_chatbot --interactive\n```\n\n运行后，您可以在终端中输入消息与您的 LLM 应用对话。按 `Ctrl + C` 即可结束会话。\n\n---\n完成上述步骤后，您已成功运行了第一个 Prompt Flow。接下来您可以探索 VS Code 插件以获得可视化的流程设计体验，或查阅官方文档学习如何进行批量测试与评估。","某电商公司的算法团队正致力于开发一款能根据用户评论自动生成个性化回复的智能客服系统，以减轻人工运营压力。\n\n### 没有 promptflow 时\n- **调试黑盒化**：开发人员难以追踪 LLM 的具体交互过程，当回复质量不佳时，无法快速定位是提示词（Prompt）问题还是代码逻辑错误。\n- **评估靠人工**：缺乏自动化评估机制，每次调整提示词后，团队需人工抽样检查数百条回复，耗时耗力且标准不一。\n- **交付门槛高**：从本地原型到生产部署需要重写大量胶水代码，环境配置复杂，导致迭代周期长达数周。\n- **协作困难**：提示词版本、测试数据和代码散落在不同文档中，团队成员间难以同步最新进展，容易引发冲突。\n\n### 使用 promptflow 后\n- **全链路可观测**：利用内置的追踪功能，团队可可视化查看每一步 LLM 的输入输出及中间变量，迅速锁定并修复逻辑缺陷。\n- **自动化批量评测**：通过集成大规模测试数据集，一键运行量化评估，用数据指标（如准确性、相关性）替代主观判断，确保持续优化。\n- **无缝生产部署**：定义好的流程可直接打包部署至选定服务平台或嵌入现有代码库，将原本数周的上线时间缩短至几天。\n- **标准化协同开发**：以“流（Flow）”为核心统一管理工作流、提示词和工具，支持团队成员在统一框架下高效协作与版本管理。\n\npromptflow 通过将分散的开发环节整合为标准化的可视工作流，让团队能从繁琐的工程细节中解脱，专注于提升大模型应用的核心质量与落地效率。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fmicrosoft_promptflow_a9760c83.png","microsoft","Microsoft","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fmicrosoft_4900709c.png","Open source projects and samples from Microsoft",null,"opensource@microsoft.com","OpenAtMicrosoft","https:\u002F\u002Fopensource.microsoft.com","https:\u002F\u002Fgithub.com\u002Fmicrosoft",[82,86,90,94,98,102,105,107,111,114],{"name":83,"color":84,"percentage":85},"Python","#3572A5",70.8,{"name":87,"color":88,"percentage":89},"HTML","#e34c26",28.3,{"name":91,"color":92,"percentage":93},"Jinja","#a52a22",0.4,{"name":95,"color":96,"percentage":97},"Jupyter Notebook","#DA5B0B",0.3,{"name":99,"color":100,"percentage":101},"Shell","#89e051",0.1,{"name":103,"color":104,"percentage":101},"PowerShell","#012456",{"name":106,"color":76,"percentage":101},"Rich Text Format",{"name":108,"color":109,"percentage":110},"Dockerfile","#384d54",0,{"name":112,"color":113,"percentage":110},"Batchfile","#C1F12E",{"name":115,"color":116,"percentage":110},"VBScript","#15dcdc",11089,1092,"2026-04-07T02:01:02","MIT","未说明",{"notes":123,"python":124,"dependencies":125},"该工具主要用于构建和评估基于 LLM 的应用流程，支持本地运行或通过 GitHub Codespaces 快速启动。可选安装 VS Code 扩展以获得图形化开发体验。若使用 Azure AI 版本需配置云端环境。默认开启遥测数据收集，可通过命令关闭。","3.9 - 3.11",[64,126],"promptflow-tools",[13,14,35,15],[129,130,131,132,133,134,135,136],"ai","llm","chatgpt","gpt","prompt","prompt-engineering","ai-application-development","ai-applications","2026-03-27T02:49:30.150509","2026-04-07T16:49:58.132630",[140,145,150,155,160,164],{"id":141,"question_zh":142,"answer_zh":143,"source_url":144},22524,"本地运行 Prompt Flow (pf test) 时出现 OpenTelemetry 属性类型错误（Invalid type NoneType）怎么办？","该问题通常由 promptflow-tracing 包与特定版本的 openai 库兼容性引起。解决方案是将所有 promptflow 相关包升级到 1.17.0 或更高版本。升级后，收集 token 指标时的空值处理逻辑已修复，不再抛出异常。执行命令：pip install --upgrade promptflow promptflow-azure promptflow-core promptflow-devkit promptflow-tracing。","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002F3751",{"id":146,"question_zh":147,"answer_zh":148,"source_url":149},22525,"使用 promptflow-vectordb 工具时遇到 BUILD_INFO 环境变量未设置的错误如何解决？","当环境中未设置 BUILD_INFO 变量时，common_index_lookup 模块会报错。临时解决方案是手动设置该环境变量。在运行 flow 之前，执行以下命令设置一个默认的 build_number：\nWindows (PowerShell): $env:BUILD_INFO='{\"build_number\":0}'\nLinux\u002FMac: export BUILD_INFO='{\"build_number\":0}'\n这将允许流程在没有实际构建信息的情况下继续运行。","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002F3837",{"id":151,"question_zh":152,"answer_zh":153,"source_url":154},22526,"如何在 Prompt Flow 中实现动态流程连接，例如 if-else 条件判断或 for-loop 循环？","目前 Prompt Flow 的节点构建基于固定的有向无环图 (DAG)，原生不支持直接的 if-else 分支或循环结构。若需实现此类动态逻辑（如 ReAct 风格提示或多步 LLM 调用），建议在单个 Python 节点中编写代码来实现循环和条件判断逻辑，而不是试图在图形界面中连接多个节点形成环路。","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002F386",{"id":156,"question_zh":157,"answer_zh":158,"source_url":159},22527,"同时使用混合搜索 (hybrid search) 和语义搜索 (semantic search) 时遇到 ParserError 或 IndexError 怎么办？","这通常是由于 YAML 配置格式错误或运行时镜像版本过旧导致的。首先检查 YAML 文件中的缩进和语法（特别是 api_base 和 deployment 字段）。如果配置无误，尝试使用更新的运行时镜像。维护者曾提供测试镜像 adramapfdev.azurecr.io\u002Fpromptflow-runtime:20240314 修复此类问题，建议将工作区自定义环境更新为包含最新修复的运行时版本。","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues\u002F2046",{"id":161,"question_zh":162,"answer_zh":163,"source_url":144},22528,"遇到 OpenAI 指标收集警告但不想禁用追踪 (tracing) 功能，有什么变通方法？","如果无法立即升级包，可以修改源码作为临时变通。在 promptflow\u002Ftracing\u002F_openai_utils.py 文件的 _get_openai_metrics_for_signal_api 方法中，添加代码删除导致类型错误的复合指标键（如 'prompt_tokens_details' 和 'completion_tokens_details'）。代码示例：\nif isinstance(usage, dict):\n    try:\n        del usage['prompt_tokens_details']\n        del usage['completion_tokens_details']\n    except KeyError:\n        pass\n但这会隐藏部分详细指标，长期方案仍是升级官方包。",{"id":165,"question_zh":166,"answer_zh":167,"source_url":154},22529,"Python 节点中如何查看 LLM 调用的中间步骤（如多轮对话或 Agent 规划过程）？","标准的 LLM 节点仅记录输入和最终输出，不显示中间 HTTP 调用细节。若需在 Python 节点中通过循环或 Agent 逻辑（如 Semantic Kernel 的 SequentialPlanner）执行多次 LLM 调用并查看中间步骤，需要在 Python 代码中显式地打印日志或将中间结果赋值给输出变量，以便在调试时进行检查。目前框架层面尚未自动展开这些内部步骤。",[169,174,179,184,189,194,199,204,209,214,219,224,229,234,239,244,249,254,259,264],{"id":170,"version":171,"summary_zh":172,"released_at":173},136217,"promptflow_1.17.1","我们很高兴地宣布 PromptFlow 1.17.1 版本正式发布。\n\n本次发布包含一些新功能、错误修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.17.1\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本将通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如遇任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 上提交报告。\n\n感谢所有为本次发布做出贡献的开发者！","2025-01-09T21:00:17",{"id":175,"version":176,"summary_zh":177,"released_at":178},136218,"promptflow_1.17.0","我们很高兴地宣布 PromptFlow 1.17.0 版本正式发布。\n\n本次发布包含多项新功能、错误修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的详细列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.17.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本现已通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如在使用过程中遇到任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 中提交报告。\n\n感谢所有为本次发布做出贡献的开发者！","2025-01-06T20:04:04",{"id":180,"version":181,"summary_zh":182,"released_at":183},136219,"promptflow_1.16.2","我们很高兴地宣布 PromptFlow 1.16.2 版本正式发布。\n\n本次发布包含多项新功能、问题修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的详细列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.16.2\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本现已通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如在使用过程中遇到任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 中提交报告。\n\n感谢所有为本次发布做出贡献的开发者！","2024-11-25T21:07:56",{"id":185,"version":186,"summary_zh":187,"released_at":188},136220,"promptflow_1.16.1","我们很高兴地宣布 PromptFlow 1.16.1 版本正式发布。\n\n本次发布包含多项新功能、错误修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的详细列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.16.1\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本现已通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如在使用过程中遇到任何问题，请前往 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 提交反馈。\n\n感谢所有为本次发布做出贡献的开发者！","2024-10-08T18:55:50",{"id":190,"version":191,"summary_zh":192,"released_at":193},136221,"promptflow_1.16.0","我们很高兴地宣布 PromptFlow 1.16.0 版本正式发布。\n\n本次发布包含多项新功能、错误修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的详细列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.16.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本现已通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如在使用过程中遇到任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 中提交反馈。\n\n感谢所有为本次发布做出贡献的开发者！","2024-09-30T16:42:02",{"id":195,"version":196,"summary_zh":197,"released_at":198},136222,"promptflow_1.15.0.post1","我们很高兴地宣布 PromptFlow 1.15.0.post1 版本正式发布。\n\n本次发布包含多项新功能、问题修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的详细列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.15.0.post1\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本已通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如在使用过程中遇到任何问题，请前往 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 提交反馈。\n\n感谢所有为本次发布做出贡献的开发者！","2024-09-23T18:57:51",{"id":200,"version":201,"summary_zh":202,"released_at":203},136223,"promptflow_1.15.0","我们很高兴地宣布 PromptFlow 1.15.0 版本正式发布。\n\n本次发布包含一些新功能、错误修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.15.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本将通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如遇任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 上提交。\n\n感谢所有为本次发布做出贡献的开发者。","2024-08-15T05:56:00",{"id":205,"version":206,"summary_zh":207,"released_at":208},136224,"promptflow_1.14.0","我们很高兴地宣布 PromptFlow 1.14.0 版本正式发布。\n\n本次发布包含一些新功能、错误修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.14.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本将通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如遇任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 上提交。\n\n感谢所有为本次发布做出贡献的开发者！","2024-07-26T02:57:50",{"id":210,"version":211,"summary_zh":212,"released_at":213},136225,"promptflow_1.12.0","我们很高兴地宣布 PromptFlow 1.12.0 版本正式发布。\n\n本次发布包含多项新功能、问题修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的详细列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.12.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本现已通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如在使用过程中遇到任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 中提交报告。\n\n感谢所有为本次发布做出贡献的开发者！","2024-06-11T08:58:23",{"id":215,"version":216,"summary_zh":217,"released_at":218},136226,"promptflow_1.11.0","我们很高兴地宣布 PromptFlow 1.11.0 版本正式发布。\n\n本次发布包含多项新功能、错误修复和改进。我们建议所有用户升级到此版本。\n\n有关所有更改的详细列表，请参阅 [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.11.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md)。\n\n该版本现已通过 PyPI 提供：\n\n```bash\npip install --upgrade promptflow\n```\n\n如在使用过程中遇到任何问题，请在 [PromptFlow 问题跟踪器](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues) 上提交反馈。\n\n感谢所有为本次发布做出贡献的开发者！","2024-05-17T07:13:07",{"id":220,"version":221,"summary_zh":222,"released_at":223},136227,"promptflow_1.10.1","We are pleased to announce the release of promptflow 1.10.1.\n\nThis release includes some new features, bug fixes, and improvements. We recommend that all users upgrade to this version.\n\nSee the [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.10.1\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md) for a list of all the changes.\n\nThe release will be available via PyPI:\n\n```bash\npip install --upgrade promptflow\n```\n\nPlease report any issues with the release on the [promptflow issue tracker](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues).\n\nThanks to all the contributors who made this release possible.\n","2024-05-10T08:29:44",{"id":225,"version":226,"summary_zh":227,"released_at":228},136228,"promptflow_1.10.0","We are pleased to announce the release of promptflow 1.10.0.\n\nThis release includes some new features, bug fixes, and improvements. We recommend that all users upgrade to this version.\n\nSee the [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.10.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md) for a list of all the changes.\n\nThe release will be available via PyPI:\n\n```bash\npip install --upgrade promptflow\n```\n\nPlease report any issues with the release on the [promptflow issue tracker](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues).\n\nThanks to all the contributors who made this release possible.\n","2024-04-26T14:28:24",{"id":230,"version":231,"summary_zh":232,"released_at":233},136229,"promptflow_1.9.0","We are pleased to announce the release of promptflow 1.9.0.\n\nThis release includes some new features, bug fixes, and improvements. We recommend that all users upgrade to this version.\n\nSee the [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.9.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md) for a list of all the changes.\n\nThe release will be available via PyPI:\n\n```bash\npip install --upgrade promptflow\n```\n\nPlease report any issues with the release on the [promptflow issue tracker](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues).\n\nThanks to all the contributors who made this release possible.\n","2024-04-17T11:06:50",{"id":235,"version":236,"summary_zh":237,"released_at":238},136230,"promptflow_1.8.0","We are pleased to announce the release of promptflow 1.8.0.\n\nThis release includes some new features, bug fixes, and improvements. We recommend that all users upgrade to this version.\n\nSee the [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.8.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md) for a list of all the changes.\n\nThe release will be available via PyPI:\n\n```bash\npip install --upgrade promptflow\n```\n\nPlease report any issues with the release on the [promptflow issue tracker](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues).\n\nThanks to all the contributors who made this release possible.\n","2024-04-10T11:57:28",{"id":240,"version":241,"summary_zh":242,"released_at":243},136231,"v1.4.0-tools","# **What's Changed**\nba9df144340f3e58a7d0559b563df057ddfaba07 Update promptflow-tools version from 1.3.0 to 1.4.0 (#2434) @chenslucky\ndd5f2d5637f6893abb0e16922261441df49e2389 Skipping test_open_model_llm.TestOpenModelLLM tests as they require new test resources (#2377) @Young Park\n7d921ad08d7c59fc78c663ce948e4aad8d3be624 [BugFix] Do not dynamic list azure deployment names for aoai vision tool in local (#2305) @Yao\n39039cd520c8fa91f754560f9b91ab454d82c645 [BugFix] set default values to workspace triad params of dynamic list tool func to avoid misleading errors (#2310) @chjinche\n77290c1f68f2a8703eb6efdeca94d64fc416d297 Improve error message when LLM tool meets gpt-4-vision-preview model (#2211) @chenslucky\nb9b72690bc5c6ba6c541bc13fc736745104d3c1a [Internal] improve token connection error message (#2245) @melionel\ne2315d665ad898768d79ea22aefea04d68190b07 update tool package changelog to latest version (#2251) @chjinche\n0eb8e0e1a4b17b4d8e28a5dc771efb757d39e6cf [NewFeature] add seed to gptv tool yamls (#2209) @chjinche\ncde5d60f8d629f434d6e52c7938ffeb285abd0f1 [NewFeature] add \"seed\" to llm tool inputs (#2207) @chjinche\n8add0988b5405af991ea5c2a6e91f9a7c2e3cc0e [NewFeature] enable AOAI token based auth for built in tool (#2153) @melionel\n07c1d9ac889368ffae162d01a2d8f34faf63a63c [BugFix]Update list_deployment_names, check connection type (#2169) @liuyuhang13 \n \n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fcompare\u002Fv1.3.0-tools...v1.4.0-tools\n","2024-03-26T10:13:11",{"id":245,"version":246,"summary_zh":247,"released_at":248},136232,"promptflow_1.7.0","We are pleased to announce the release of promptflow 1.7.0.\n\nThis release includes some new features, bug fixes, and improvements. We recommend that all users upgrade to this version.\n\nSee the [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.7.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md) for a list of all the changes.\n\nThe release will be available via PyPI:\n\n```bash\npip install --upgrade promptflow\n```\n\nPlease report any issues with the release on the [promptflow issue tracker](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues).\n\nThanks to all the contributors who made this release possible.\n","2024-03-25T02:47:56",{"id":250,"version":251,"summary_zh":252,"released_at":253},136233,"promptflow_1.6.0","We are pleased to announce the release of promptflow 1.6.0.\n\nThis release includes some new features, bug fixes, and improvements. We recommend that all users upgrade to this version.\n\nSee the [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.6.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md) for a list of all the changes.\n\nThe release will be available via PyPI:\n\n```bash\npip install --upgrade promptflow\n```\n\nPlease report any issues with the release on the [promptflow issue tracker](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues).\n\nThanks to all the contributors who made this release possible.\n","2024-03-01T09:34:19",{"id":255,"version":256,"summary_zh":257,"released_at":258},136234,"v1.3.0-tools","# **What's Changed**\r\nc210b024499140c5d5daca4b3444f015106a7f75 Upgrade tool version to 1.3.0 (#2112) @Yao\r\nce46f3848b6594f595eb0b89deef5dde719a94b0 [Bugfix] Update expected error message of 'test_completion_with_chat_model' (#2115) @chjinche\r\n71b187c4dc51d79307c37f0e74c180803eaa53af [Internal] Refine the retry-after interval for openai retry error (#2073) @Ge Gao\r\n83771ccfd1446c7640a5d35251229b21757ba389 [Bugfix] fix naming error of legacy 'OpenSourceLLMOnlineEndpointError' (#2059) @chjinche\r\nf36455c964a43d6b734ff53be918e9be20460f2f disable openai built-in retry mechanism for better debuggability and real-time status updates (#1769) @chjinche \r\n \r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fcompare\u002Fv1.2.0-tools...v1.3.0-tools\r\n","2024-03-01T10:08:22",{"id":260,"version":261,"summary_zh":262,"released_at":263},136235,"v1.2.0-tools","# **What's Changed**\nd701cb04ecf2cb70e13245d8d88657ff3ca592ba Upgrade tool version to 1.2.0 (#1981) @chenslucky\ne9e6fc3e767969345bdae87c18b7b0fa5ccb6025 [Tools] Support Azure OpenAI GPT-4 Turbo with Vision auto list deployment_name (#1788) @liuyuhang13\n122d498400901d773f8bdcf5e064f8b8fe8c208f Match the promptflow-tools setup to promptflow setup (#1798) @Hannah K \n \n**Full Changelog**: https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fcompare\u002Fv1.1.0-tools...v1.2.0-tools\n","2024-02-07T02:40:43",{"id":265,"version":266,"summary_zh":267,"released_at":268},136236,"promptflow_1.5.0","We are pleased to announce the release of promptflow 1.5.0.\n\nThis release includes some new features, bug fixes, and improvements. We recommend that all users upgrade to this version.\n\nSee the [CHANGELOG](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fblob\u002Frelease\u002Fpromptflow\u002F1.5.0\u002Fsrc\u002Fpromptflow\u002FCHANGELOG.md) for a list of all the changes.\n\nThe release will be available via PyPI:\n\n```bash\npip install --upgrade promptflow\n```\n\nPlease report any issues with the release on the [promptflow issue tracker](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpromptflow\u002Fissues).\n\nThanks to all the contributors who made this release possible.\n","2024-02-06T10:56:28"]