[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-retkowsky--Azure-AIGEN-demos":3,"tool-retkowsky--Azure-AIGEN-demos":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},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。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},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",138956,2,"2026-04-05T11:33:21",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},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。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":80,"owner_email":81,"owner_twitter":82,"owner_website":83,"owner_url":84,"languages":85,"stars":97,"forks":98,"last_commit_at":99,"license":82,"difficulty_score":23,"env_os":100,"env_gpu":100,"env_ram":100,"env_deps":101,"category_tags":104,"github_topics":105,"view_count":23,"oss_zip_url":82,"oss_zip_packed_at":82,"status":16,"created_at":123,"updated_at":124,"faqs":125,"releases":126},2785,"retkowsky\u002FAzure-AIGEN-demos","Azure-AIGEN-demos","Azure AI Foundry (demos, documentation, accelerators). ","Azure-AIGEN-demos 是微软官方推出的 Azure AI Foundry 实战资源库，汇集了丰富的演示代码、技术文档和加速开发模板。它旨在解决企业在构建生成式 AI 应用时面临的痛点：如何快速整合多样化的模型（如 GPT-5、Mistral、Cohere、Flux 等）与工具，同时无需耗费大量精力管理底层基础设施。\n\n通过提供开箱即用的示例，Azure-AIGEN-demos 帮助开发者直接在统一平台上实现智能体编排、文档分析、图像异常检测、自动打标及实时交互等复杂场景。其核心亮点在于展示了如何利用 Azure AI Foundry 的企业级能力，包括内置的链路追踪、监控评估、基于角色的访问控制（RBAC）以及标准化的网络策略，让团队能专注于业务逻辑创新而非运维细节。\n\n这套资源特别适合 AI 应用开发者、架构师及技术研究人员使用。无论你是希望快速验证新模型效果，还是需要将实验性项目转化为具备生产就绪能力的企业级应用，Azure-AIGEN-demos 都提供了从概念验证到落地部署的全流程参考，助力用户高效驾驭微软统一的 AI 开发生态。","## Microsoft Foundry\n\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fretkowsky_Azure-AIGEN-demos_readme_8ea30a2d3b6a.jpg\">\n\u003Cbr>\nMicrosoft Foundry is a unified Azure platform-as-a-service offering for enterprise AI operations, model builders, and application development. This foundation combines production-grade infrastructure with friendly interfaces, enabling developers to focus on building applications rather than managing infrastructure.\u003Cbr>\u003Cbr>\nMicrosoft Foundry unifies agents, models, and tools under a single management grouping with built-in enterprise-readiness capabilities including tracing, monitoring, evaluations, and customizable enterprise setup configurations. The platform provides streamlined management through unified Role-based access control (RBAC), networking, and policies under one Azure resource provider namespace.\n\u003Cbr>\u003Cbr>\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fretkowsky_Azure-AIGEN-demos_readme_5edd7d46726a.jpg\">\n\u003Cbr>\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fretkowsky_Azure-AIGEN-demos_readme_515e55187f29.jpg\">\n\u003Cbr>\n\n[Microsoft Foundry portal](https:\u002F\u002Fai.azure.com\u002F)\n\n\u003Cbr>\n\n---\n\n## Latest Content\n\n### New content (19 February 2026)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 mistral-document-ai-2512| mistral-document-ai-2512 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Ftree\u002Fmain\u002Fmistral-document-ai-2512\n\n### New content (16 February 2026)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Image Anomaly Detection with Cohere Embed 4 on Azure AI Foundry | Image Anomaly Detection| https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fanomaly_detection_cohere_embed4\u002FImage%20Anomaly%20Detection%20with%20Cohere%20Embed%204%20on%20Azure%20AI%20Foundry.ipynb\n| 🔥 Auto-Tagging Images with Cohere Embed 4 on Microsoft Foundry | Auto tagging | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fauto_tagging_cohere\u002Fauto_tagging_cohere_embed4_azure.ipynb\n\n### New content (06 February 2026)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 gpt-5.2 models | gpt-5.2 examples | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fgpt-5.2\u002Fgpt52_models.ipynb|\n\n### New content (02 February 2026)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Azure Prices Fetcher | Azure API pricing. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002FPricing\u002Fazure_prices_fetcher.ipynb |\n\n\n### New content (27 January 2026)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 gpt-realtime-mini | gpt-realtime-mini with Microsoft Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fgpt-realtime-mini\u002Fgpt_realtime_mini_azure.ipynb |\n\n\n### New content (16 January 2026)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Azure AI Agent MCP | Azure AI Agent with MCP connexion. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002FMCP\u002FMCP_Microsoft_Learn_Chatbot.ipynb |\n\n\n### New content (09 September 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Mistral Document AI | End‑to‑end examples of Mistral Document AI within Azure AI Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FMistral%20Document%20AI\u002FMistral%20Document%20AI%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Flux.1 Kontext Pro – Text & Image‑to‑Image | Image editing scenarios using Flux.1 Kontext Pro with Azure AI Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FFlux.1%20Kontext%20Pro\u002FImage%20Edition%20with%20Flux.1%20Kontext%20Pro%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Flux1.1 Pro – Text‑to‑Image | High‑quality text‑to‑image generation using FLUX‑1.1‑pro in Azure AI Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fblackforestslabs\u002Fflux1.1pro\u002FText%20to%20image%20with%20FLUX-1.1-pro%20in%20Azure%20AI%20Foundry.ipynb |\n\n### New content (26 August 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 GPT‑5 demo examples | GPT‑5 usage patterns and reference scenarios in Azure AI Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt5\u002FAzure%20AI%20Foundry%20-%20gpt5.ipynb |\n\n### New content (26 June 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Azure AI Agent Service – Bing integration (update) | Updated Bing search integration with Azure AI Agent Service. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F6%20Azure%20AI%20Agent%20service%20-%20Bing%20integration.ipynb |\n| 🔥 Azure AI Agent Service – Custom Bing integration | Customizable Bing‑based agents with Azure AI Agent Service. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F7%20Azure%20AI%20Agent%20service%20-%20Custom%20Bing%20agent.ipynb |\n| 🔥 Azure AI Agent Service – Connected agents | Patterns for orchestrating multiple connected agents. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F8%20Azure%20AI%20Agent%20service%20-%20Connected%20agents.ipynb |\n| 🔥 Grok with Azure AI Foundry | Integration scenarios using Grok models within Azure AI Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FGrok\u002FGrok%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Phi‑4 reasoning with Azure AI Foundry | Advanced reasoning workflows using Phi‑4 in Azure AI Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fphi-4%20reasoning\u002FPhi-4%20reasoning%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Generative AI model tracing with Azure AI Foundry | Observability and tracing for generative AI workloads. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Ftracing\u002FAzure%20AI%20Foundry%20tracing.ipynb |\n| 🔥 Agents evaluator with Azure AI Foundry | Evaluation of agentic workflows and behaviors. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FAgents%20evaluators.ipynb |\n| 🔥 Azure OpenAI evaluators with Azure AI Foundry | Built‑in evaluators for Azure OpenAI workloads. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FAzure%20OpenAI%20evaluators.ipynb |\n| 🔥 Evaluators with Azure AI Foundry | General evaluation flows for generative applications. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FEvaluators.ipynb |\n| 🔥 Custom evaluators with Azure AI Foundry | Authoring and integrating custom evaluation logic. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FCustom%20evaluators.ipynb |\n| 🔥 Retrieval evaluators with Azure AI Foundry | Quality evaluation for retrieval and RAG scenarios. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FRetrieval%20evaluators.ipynb |\n| 🔥 Risk and safety evaluators with Azure AI Foundry | Risk, safety, and compliance‑oriented evaluation flows. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FRisk%20and%20safety%20evaluators%20with%20Azure%20AI%20Foundry.ipynb |\n\n### New content (02 June 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 SORA with Azure AI Foundry | End‑to‑end SORA integration scenarios. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fsora\u002FSORA%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Image‑to‑video with GPT‑4o and SORA | Image‑to‑video generation pipeline using GPT‑4o and SORA. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fsora\u002FImage%20to%20Video%20with%20gpt4o%20and%20SORA.ipynb |\n| 🔥 Video‑to‑video with GPT‑4o and SORA | Video transformation workflows combining GPT‑4o and SORA. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fsora\u002FVideo%20to%20Video%20with%20gpt4o%20and%20SORA.ipynb |\n| 🔥 Agentic retrieval in Azure AI Search | Practical scenarios for agentic retrieval using Azure AI Search. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAgentic%20RAG\u002FIntroducing%20agentic%20retrieval%20in%20Azure%20AI%20Search.ipynb |\n| 🔥 Model router | Routing requests dynamically across multiple models. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FModel%20router\u002FModel%20router.ipynb |\n\n### New content (21 May 2025)\n\n**AutoGen series**\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 AutoGen – Settings | Configuration patterns and best practices for AutoGen. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Introduction | Conceptual and architectural introduction to AutoGen. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Simple agent for financial analysis | Scenario using AutoGen agents for financial data analysis. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Azure AI Agent integration | Integration of AutoGen with Azure AI Agent Service. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Chatbot | Chat‑oriented agent implementation with AutoGen. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Enabling LLM‑powered agents to cooperate | Coordinating multiple agents collaboratively. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Multi‑agents | Multi‑agent orchestration patterns. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Multi‑agent with image generation | Multi‑agent workflows integrating image generation. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Human interaction | Human‑in‑the‑loop interactions within AutoGen flows. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – Multimodal | Multimodal scenarios (text, image, etc.) with AutoGen. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n\n### New content (30 April 2025)\n\n**Azure AI Agent Service**\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Single‑agent pattern | Basic, single‑agent implementation with Azure AI Agent Service. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F1%20Azure%20AI%20Agent%20service%20-%20Single%20agent.ipynb |\n| 🔥 Multi‑agent orchestration | Coordination of several agents for complex workflows. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F2%20Azure%20AI%20Agent%20service%20-%20Many%20agents.ipynb |\n| 🔥 File search (simple RAG) | Simple retrieval‑augmented generation over files. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F3%20Azure%20AI%20Agent%20Service%20-%20File%20Search.ipynb |\n| 🔥 Code interpreter (EDA on a dataset) | Exploratory data analysis using the code interpreter tool. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F4%20Azure%20AI%20Agent%20service%20-%20Code%20interpreter.ipynb |\n| 🔥 User function (Azure Maps Weather Services) | Function‑calling integration with Azure Maps Weather Services. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F5%20Azure%20AI%20Agent%20service%20-%20Function%20calling.ipynb |\n| 🔥 Bing Search integration | Using Bing Search as a tool from agents. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F6%20Azure%20AI%20Agent%20service%20-%20Bing%20integration.ipynb |\n\n### New content (29 April 2025)\n\n**GPT‑image‑1 on Azure AI Foundry**\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Image generation | Text‑to‑image generation scenarios with gpt‑image‑1. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20generation.ipynb |\n| 🔥 Image editing | Image editing workflows based on existing images. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20edition.ipynb |\n| 🔥 Image composition | Composing multiple elements into a single generated image. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20Compose.ipynb |\n| 🔥 Image inpainting | Inpainting and localized image modification scenarios. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20Inpainting.ipynb |\n\n### New content (18 April 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Mistral in Azure AI Foundry | General Mistral model usage in Azure AI Foundry. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fmistral\u002Fmistral.ipynb |\n| 🔥 Mistral OCR in Azure AI Foundry | OCR‑oriented scenarios using Mistral. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fmistral\u002Fmistral%20OCR.ipynb |\n| 🔥 o1 on images | Image‑centric reasoning and analysis using the o1 model. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo1\u002Fo1%20on%20images.ipynb |\n| 🔥 Stored completions with Azure AI Foundry | Using stored completions to optimize performance and cost. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FStored%20completions\u002FStored%20completions.ipynb |\n| 🔥 Responses API examples | Core usage patterns of the Responses API. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FResponses%20API\u002FResponses%20API%20examples.ipynb |\n| 🔥 Responses API web app | Web application example built on top of the Responses API. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FResponses%20API\u002FResponses%20API%20webapp.ipynb |\n| 🔥 GPT‑4.1 examples | Reference workflows for GPT‑4.1 in Azure OpenAI. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt41\u002Fgpt41.ipynb |\n| 🔥 gpt‑4o mini TTS | Text‑to‑speech scenarios with gpt‑4o mini. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt4ominiTTS\u002Fgpt4ominiTTS.ipynb |\n| 🔥 gpt‑4o mini transcription | Speech‑to‑text and transcription examples with gpt‑4o mini. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt4ominitranscribe\u002Fgpt-4o%20mini%20transcribe.ipynb |\n| 🔥 o4‑mini examples | Text‑focused o4‑mini usage scenarios. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo4mini\u002FAzure%20OpenAI%20o4%20mini%20examples.ipynb |\n| 🔥 o4‑mini on images | Image‑based workflows with o4‑mini. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo4mini\u002Fo4%20mini%20examples%20on%20images.ipynb |\n\n### New content (14 February 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 o1‑mini | Compact, cost‑efficient reasoning with o1‑mini. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo1\u002FAzure%20OpenAI%20o1%20mini%20examples.ipynb |\n| 🔥 o3‑mini | Advanced lightweight reasoning with o3‑mini. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo3\u002FAzure%20OpenAI%20o3%20mini%20examples.ipynb |\n| 🔥 GPT‑4o fine‑tuning (text) | Text classification with a fine‑tuned GPT‑4o model. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FGpt-4o-Text-FineTuning\u002FText%20classification%20with%20gpt-4o%20fine%20tuned%20model.ipynb |\n\n### New content (06 February 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Azure OpenAI audio generation | Audio generation flows using GPT‑4o. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20OpenAI%20audio%20generation\u002FAzure%20OpenAI%20Gpt4o%20Audio.ipynb |\n\n### New content (23 January 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 Image classification with gpt‑4o | Baseline image classification with GPT‑4o. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002Fgpt-4o-image-classification |\n| 🔥 gpt‑4o model fine‑tuning for image classification | Fine‑tuning GPT‑4o for industrial image classification (NEU dataset). | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002Fgpt-4o-image-classification-finetuning |\n\n### New content (16 January 2025)\n\n| Item | Description | Link |\n| --- | --- | --- |\n| 🔥 AI audio and video podcast generator | Automated podcast production with Azure OpenAI, Azure AI Document Intelligence, and Azure AI Speech. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAI%20podcast%20generation |\n| 🔥 GPT‑4o fine‑tuning for VQA | Visual question answering using a fine‑tuned GPT‑4o model. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FGpt-4o%20Fine%20tuning |\n\n---\n\n## Azure OpenAI Demos – Thematic Overview\n\n| Area | Description | Link |\n| --- | --- | --- |\n| Azure OpenAI basics | Introductory scenarios for getting started with Azure OpenAI. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FBasics |\n| Azure OpenAI quick demos | Short, workshop‑oriented samples. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAzure%20Open%20AI%20quick%20demos |\n| Vector embeddings | Embeddings for text, images, and audio. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FEmbeddings |\n| Embeddings with pandas | Embedding patterns over pandas data frames. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FEmbeddings%20with%20Pandas |\n| Azure Computer Vision and LangChain | Using Azure Computer Vision with LangChain. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAzure%20Computer%20Vision%20and%20Langchain |\n| Azure Cognitive Search – vector search & JSON | Vector search and JSON document analysis with Azure OpenAI. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAzure%20Cognitive%20Search%20Vector%20Search%20Code%20Sample%20with%20Azure%20OpenAI |\n| Python code analysis | Analysis of Python notebooks with LangChain and Azure Cognitive Search. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FCode%20analysis%20with%20Langchain%20%2B%20Azure%20OpenAI%20%2B%20Azure%20Cognitive%20Search%20(vector%20store) |\n| PDF document analysis | PDF analysis workflows using LangChain, Azure OpenAI, and Azure Cognitive Search. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FLanchain%20with%20Azure%20Open%20AI%20(PDF%20files)%20and%20Azure%20Cognitive%20Search |\n| LLaMA | Introductory LLaMA‑based scenarios. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FLlama |\n| DALL‑E 2 image generation | Image generation with DALL‑E 2 in Azure OpenAI. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FArtificial%20images%20with%20Dall-e%202 |\n| Python function integration | Function calling and Python function orchestration. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FPython%20functions%20integration |\n| Video Indexer analysis | Analysing Azure Video Indexer transcripts with Azure OpenAI. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FVideo%20Indexer%20analysis |\n| Email response generation | Intelligent reply generation for email content. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FEmail%20response%20generation |\n| Wikification | Entity‑centric enrichment and wikification flows. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FWikification |\n| Resume analysis | CV\u002Fresume parsing, extraction, and scoring. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FResume%20analysis |\n| Text analytics & sentiment | Text analytics and sentiment analysis with Azure OpenAI. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FText%20analytics%20with%20Azure%20Open%20AI |\n| Prompt Flow model invocation | Calling deployed Prompt Flow models from code. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FPromptFlow%20model%20deployment |\n| From text to emojis | Emoji‑based categorisation of text. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FFrom%20text%20to%20emoji |\n| Code optimization and conversion | Refactoring, optimisation, and language conversion of code. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FCode%20Optimization%20and%20conversion |\n| PowerPoint generation | Automatic PowerPoint generation with Azure OpenAI. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FPowerPoint%20generation%20with%20Azure%20Open%20AI |\n| FHIR analysis | Healthcare FHIR data analysis scenarios. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FFHIR%20analysis |\n| Utilities | Reusable utilities for Azure OpenAI projects. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FUtilities |\n| Meeting audio analysis | End‑to‑end flow for meeting audio analysis with Azure OpenAI and Azure Speech. | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAnalyse%20audio%20meeting%20notes%20with%20Azure%20Open%20AI%20and%20Azure%20Speech%20Services |\n\n---\n\n## Documentation\n\n| Item | Description | Link |\n| --- | --- | --- |\n| Microsoft Foundry – product page | Product overview, capabilities, and pricing. | https:\u002F\u002Fazure.microsoft.com\u002Fen-us\u002Fproducts\u002Fai-foundry\u002F#AI-Foundry-Hero |\n| What is Microsoft Foundry? | Conceptual documentation and key architectural concepts. | https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fazure\u002Fai-foundry\u002Fwhat-is-azure-ai-foundry |\n\n---\n\n## Author\n\n| Field | Details |\n| --- | --- |\n| Name | Serge Retkowsky |\n| Created | 05 September 2023 |\n| Last updated | 19 February 2026 |\n| Email | serge.retkowsky@microsoft.com |\n| LinkedIn | https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fserger\u002F |\n| Medium publications | https:\u002F\u002Fmedium.com\u002F@sergems18\u002F |\n","## 微软 Foundry\n\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fretkowsky_Azure-AIGEN-demos_readme_8ea30a2d3b6a.jpg\">\n\u003Cbr>\n微软 Foundry 是面向企业 AI 运营、模型构建者和应用开发的统一 Azure 平台即服务产品。这一基础架构将生产级基础设施与友好的用户界面相结合，使开发者能够专注于构建应用程序，而非管理基础设施。\u003Cbr>\u003Cbr>\n微软 Foundry 将智能体、模型和工具统一在一个管理框架下，并内置了追踪、监控、评估以及可自定义的企业级部署配置等企业就绪功能。该平台通过统一的基于角色的访问控制 (RBAC)、网络和策略，在一个 Azure 资源提供程序命名空间内实现简化的管理。\n\u003Cbr>\u003Cbr>\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fretkowsky_Azure-AIGEN-demos_readme_5edd7d46726a.jpg\">\n\u003Cbr>\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fretkowsky_Azure-AIGEN-demos_readme_515e55187f29.jpg\">\n\u003Cbr>\n\n[微软 Foundry 门户](https:\u002F\u002Fai.azure.com\u002F)\n\n\u003Cbr>\n\n---\n\n## 最新内容\n\n### 新内容（2026年2月19日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 mistral-document-ai-2512| mistral-document-ai-2512 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Ftree\u002Fmain\u002Fmistral-document-ai-2512\n\n### 新内容（2026年2月16日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 使用 Cohere Embed 4 在 Azure AI Foundry 上进行图像异常检测 | 图像异常检测| https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fanomaly_detection_cohere_embed4\u002FImage%20Anomaly%20Detection%20with%20Cohere%20Embed%204%20on%20Azure%20AI%20Foundry.ipynb\n| 🔥 使用 Cohere Embed 4 在 Microsoft Foundry 上自动标记图像 | 自动标记 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fauto_tagging_cohere\u002Fauto_tagging_cohere_embed4_azure.ipynb\n\n### 新内容（2026年2月6日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 gpt-5.2 模型 | gpt-5.2 示例 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fgpt-5.2\u002Fgpt52_models.ipynb|\n\n### 新内容（2026年2月2日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 Azure 价格获取器 | Azure API 价格查询。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002FPricing\u002Fazure_prices_fetcher.ipynb |\n\n### 新内容（2026年1月27日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 gpt-realtime-mini | gpt-realtime-mini 结合 Microsoft Foundry 使用。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002Fgpt-realtime-mini\u002Fgpt_realtime_mini_azure.ipynb |\n\n### 新内容（2026年1月16日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 Azure AI 智能体 MCP | 带有 MCP 连接的 Azure AI 智能体。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos\u002Fblob\u002Fmain\u002FMCP\u002FMCP_Microsoft_Learn_Chatbot.ipynb |\n\n### 新内容（2025年9月9日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 Mistral 文档 AI | 在 Azure AI Foundry 内使用 Mistral 文档 AI 的端到端示例。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FMistral%20Document%20AI\u002FMistral%20Document%20AI%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Flux.1 Kontext Pro – 文本与图像生成 | 使用 Flux.1 Kontext Pro 结合 Azure AI Foundry 的图像编辑场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FFlux.1%20Kontext%20Pro\u002FImage%20Edition%20with%20Flux.1%20Kontext%20Pro%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Flux1.1 Pro – 文本转图像 | 在 Azure AI Foundry 中使用 FLUX‑1.1‑pro 实现高质量的文本到图像生成。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fblackforestslabs\u002Fflux1.1pro\u002FText%20to%20image%20with%20FLUX-1.1-pro%20in%20Azure%20AI%20Foundry.ipynb |\n\n### 新内容（2025年8月26日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 GPT‑5 演示示例 | 在 Azure AI Foundry 中 GPT‑5 的使用模式及参考场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt5\u002FAzure%20AI%20Foundry%20-%20gpt5.ipynb |\n\n### 新内容（2025年6月26日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 Azure AI 智能体服务 – Bing 集成（更新） | 更新后的 Bing 搜索集成与 Azure AI 智能体服务。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F6%20Azure%20AI%20Agent%20service%20-%20Bing%20integration.ipynb |\n| 🔥 Azure AI 智能体服务 – 自定义 Bing 集成 | 可定制的基于 Bing 的智能体与 Azure AI 智能体服务。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F7%20Azure%20AI%20Agent%20service%20-%20Custom%20Bing%20agent.ipynb |\n| 🔥 Azure AI 智能体服务 – 连接的智能体 | 多个连接智能体的编排模式。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F8%20Azure%20AI%20Agent%20service%20-%20Connected%20agents.ipynb |\n| 🔥 Grok 结合 Azure AI Foundry | 在 Azure AI Foundry 中使用 Grok 模型的集成场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FGrok\u002FGrok%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Phi‑4 推理结合 Azure AI Foundry | 在 Azure AI Foundry 中使用 Phi‑4 进行高级推理工作流。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fphi-4%20reasoning\u002FPhi-4%20reasoning%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 Azure AI Foundry 中的生成式 AI 模型追踪 | 用于生成式 AI 工作负载的可观测性和追踪功能。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Ftracing\u002FAzure%20AI%20Foundry%20tracing.ipynb |\n| 🔥 Azure AI Foundry 中的智能体评估器 | 用于评估智能体工作流和行为的工具。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FAgents%20evaluators.ipynb |\n| 🔥 Azure OpenAI 评估器结合 Azure AI Foundry | 专为 Azure OpenAI 工作负载设计的内置评估工具。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FAzure%20OpenAI%20evaluators.ipynb |\n| 🔥 Azure AI Foundry 中的通用评估器 | 适用于生成式应用的一般评估流程。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FEvaluators.ipynb |\n| 🔥 自定义评估器结合 Azure AI Foundry | 自定义评估逻辑的编写与集成。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FCustom%20evaluators.ipynb |\n| 🔥 检索评估器结合 Azure AI Foundry | 用于检索和 RAG 场景的质量评估。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FRetrieval%20evaluators.ipynb |\n| 🔥 风险与安全评估器结合 Azure AI Foundry | 面向风险、安全和合规的评估流程。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FObservability\u002FRisk%20and%20safety%20evaluators%20with%20Azure%20AI%20Foundry.ipynb |\n\n### 新内容（2025年6月2日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 SORA 与 Azure AI Foundry | 端到端的 SORA 集成场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fsora\u002FSORA%20with%20Azure%20AI%20Foundry.ipynb |\n| 🔥 使用 GPT‑4o 和 SORA 的图像转视频 | 基于 GPT‑4o 和 SORA 的图像转视频生成流水线。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fsora\u002FImage%20to%20Video%20with%20gpt4o%20and%20SORA.ipynb |\n| 🔥 使用 GPT‑4o 和 SORA 的视频转视频 | 结合 GPT‑4o 和 SORA 的视频转换工作流。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fsora\u002FVideo%20to%20Video%20with%20gpt4o%20and%20SORA.ipynb |\n| 🔥 Azure AI Search 中的智能体检索 | 使用 Azure AI Search 进行智能体检索的实用场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAgentic%20RAG\u002FIntroducing%20agentic%20retrieval%20in%20Azure%20AI%20Search.ipynb |\n| 🔥 模型路由 | 在多个模型之间动态路由请求。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FModel%20router\u002FModel%20router.ipynb |\n\n### 新内容（2025年5月21日）\n\n**AutoGen 系列**\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 AutoGen – 设置 | AutoGen 的配置模式和最佳实践。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 简介 | AutoGen 的概念与架构介绍。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 用于财务分析的简单智能体 | 使用 AutoGen 智能体进行财务数据分析的场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 与 Azure AI Agent 服务集成 | AutoGen 与 Azure AI Agent 服务的集成。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 聊天机器人 | 使用 AutoGen 实现面向聊天的智能体。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 使 LLM 驱动的智能体协作 | 协调多个智能体协同工作。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 多智能体 | 多智能体编排模式。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 具有图像生成功能的多智能体 | 集成图像生成的多智能体工作流。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 人机交互 | 在 AutoGen 流程中实现人工介入的交互。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n| 🔥 AutoGen – 多模态 | 使用 AutoGen 处理文本、图像等多种模态的场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAutogen |\n\n### 新内容（2025年4月30日）\n\n**Azure AI Agent 服务**\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 单智能体模式 | 使用 Azure AI Agent 服务实现的基本单智能体方案。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F1%20Azure%20AI%20Agent%20service%20-%20Single%20agent.ipynb |\n| 🔥 多智能体编排 | 协调多个智能体以完成复杂的工作流程。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F2%20Azure%20AI%20Agent%20service%20-%20Many%20agents.ipynb |\n| 🔥 文件搜索（简单 RAG） | 基于文件的简单检索增强生成。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F3%20Azure%20AI%20Agent%20Service%20-%20File%20Search.ipynb |\n| 🔥 代码解释器（数据集的探索性数据分析） | 使用代码解释器工具进行探索性数据分析。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F4%20Azure%20AI%20Agent%20service%20-%20Code%20interpreter.ipynb |\n| 🔥 用户函数（Azure Maps 天气服务） | 与 Azure Maps 天气服务的函数调用集成。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F5%20Azure%20AI%20Agent%20service%20-%20Function%20calling.ipynb |\n| 🔥 Bing 搜索集成 | 将 Bing 搜索作为智能体的工具使用。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20Agent%20Service\u002F6%20Azure%20AI%20Agent%20service%20-%20Bing%20integration.ipynb |\n\n### 新内容（2025年4月29日）\n\n**Azure AI Foundry 上的 GPT‑image‑1**\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 图像生成 | 使用 gpt‑image‑1 进行文本到图像的生成场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20generation.ipynb |\n| 🔥 图像编辑 | 基于现有图像的图像编辑工作流。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20edition.ipynb |\n| 🔥 图像合成 | 将多个元素组合成一张生成的图像。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20Compose.ipynb |\n| 🔥 图像修复 | 图像修复及局部修改场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt-image-1\u002FAzure%20AI%20Foundry%20gpt-image-1%20-%20Image%20Inpainting.ipynb |\n\n### 新内容（2025年4月18日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 Azure AI Foundry 中的 Mistral | 在 Azure AI Foundry 中通用的 Mistral 模型用法。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fmistral\u002Fmistral.ipynb |\n| 🔥 Azure AI Foundry 中的 Mistral OCR | 使用 Mistral 的 OCR 相关场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fmistral\u002Fmistral%20OCR.ipynb |\n| 🔥 o1 处理图像 | 使用 o1 模型进行以图像为中心的推理和分析。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo1\u002Fo1%20on%20images.ipynb |\n| 🔥 Azure AI Foundry 中的存储式补全 | 使用存储式补全优化性能和成本。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FStored%20completions\u002FStored%20completions.ipynb |\n| 🔥 Responses API 示例 | Responses API 的核心使用模式。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FResponses%20API\u002FResponses%20API%20examples.ipynb |\n| 🔥 Responses API Web 应用 | 基于 Responses API 构建的 Web 应用示例。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FResponses%20API\u002FResponses%20API%20webapp.ipynb |\n| 🔥 GPT‑4.1 示例 | Azure OpenAI 中 GPT‑4.1 的参考工作流。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt41\u002Fgpt41.ipynb |\n| 🔥 gpt‑4o mini 文本转语音 | 使用 gpt‑4o mini 的文本转语音场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt4ominiTTS\u002Fgpt4ominiTTS.ipynb |\n| 🔥 gpt‑4o mini 语音转文字 | 使用 gpt‑4o mini 的语音转文字和转录示例。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fgpt4ominitranscribe\u002Fgpt-4o%20mini%20transcribe.ipynb |\n| 🔥 o4‑mini 示例 | 以文本为主的 o4‑mini 使用场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo4mini\u002FAzure%20OpenAI%20o4%20mini%20examples.ipynb |\n| 🔥 o4‑mini 处理图像 | 使用 o4‑mini 进行基于图像的工作流。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo4mini\u002Fo4%20mini%20examples%20on%20images.ipynb |\n\n### 新内容（2025年2月14日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 o1‑mini | 使用 o1‑mini 进行紧凑且经济高效的推理。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo1\u002FAzure%20OpenAI%20o1%20mini%20examples.ipynb |\n| 🔥 o3‑mini | 使用 o3‑mini 进行先进的轻量级推理。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002Fo3\u002FAzure%20OpenAI%20o3%20mini%20examples.ipynb |\n| 🔥 GPT‑4o 微调（文本） | 使用微调后的 GPT‑4o 模型进行文本分类。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FGpt-4o-Text-FineTuning\u002FText%20classification%20with%20gpt-4o%20fine%20tuned%20model.ipynb |\n\n### 新内容（2025年2月6日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 Azure OpenAI 音频生成 | 使用 GPT‑4o 进行音频生成流程。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Fblob\u002Fmain\u002FAzure%20OpenAI%20audio%20generation\u002FAzure%20OpenAI%20Gpt4o%20Audio.ipynb |\n\n### 新内容（2025年1月23日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 使用 gpt‑4o 进行图像分类 | 使用 GPT‑4o 进行基础图像分类。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002Fgpt-4o-image-classification |\n| 🔥 用于图像分类的 gpt‑4o 模型微调 | 对 GPT‑4o 进行工业级图像分类（NEU 数据集）的微调。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002Fgpt-4o-image-classification-finetuning |\n\n### 新内容（2025年1月16日）\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| 🔥 AI 音频与视频播客生成器 | 使用 Azure OpenAI、Azure AI Document Intelligence 和 Azure AI Speech 自动化制作播客。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAI%20podcast%20generation |\n| 🔥 GPT‑4o 微调用于 VQA | 使用微调后的 GPT‑4o 模型进行视觉问答任务。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FGpt-4o%20Fine%20tuning |\n\n---\n\n## Azure OpenAI 演示 – 主题概述\n\n| 领域 | 描述 | 链接 |\n| --- | --- | --- |\n| Azure OpenAI 基础 | 用于开始使用 Azure OpenAI 的入门场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FBasics |\n| Azure OpenAI 快速演示 | 短小、面向研讨会的示例。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAzure%20Open%20AI%20quick%20demos |\n| 向量嵌入 | 文本、图像和音频的嵌入。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FEmbeddings |\n| 使用 Pandas 的嵌入 | 在 Pandas 数据框上进行嵌入模式操作。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FEmbeddings%20with%20Pandas |\n| Azure 计算机视觉与 LangChain | 将 Azure 计算机视觉与 LangChain 结合使用。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAzure%20Computer%20Vision%20and%20Langchain |\n| Azure 认知搜索 – 向量搜索与 JSON | 使用 Azure OpenAI 进行向量搜索和 JSON 文档分析。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAzure%20Cognitive%20Search%20Vector%20Search%20Code%20Sample%20with%20Azure%20OpenAI |\n| Python 代码分析 | 使用 LangChain 和 Azure 认知搜索分析 Python 笔记本。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FCode%20analysis%20with%20Langchain%20%2B%20Azure%20Open%20AI%20%2B%20Azure%20Cognitive%20Search%20(vector%20store) |\n| PDF 文档分析 | 使用 LangChain、Azure OpenAI 和 Azure 认知搜索的 PDF 分析工作流。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FLanchain%20with%20Azure%20Open%20AI%20(PDF%20files)%20and%20Azure%20Cognitive%20Search |\n| LLaMA | 基于 LLaMA 的入门场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FLlama |\n| DALL‑E 2 图像生成 | 在 Azure OpenAI 中使用 DALL‑E 2 进行图像生成。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FArtificial%20images%20with%20Dall-e%202 |\n| Python 函数集成 | 函数调用和 Python 函数编排。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FPython%20functions%20integration |\n| Video Indexer 分析 | 使用 Azure OpenAI 分析 Azure Video Indexer 的转录文本。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FVideo%20Indexer%20analysis |\n| 邮件回复生成 | 对电子邮件内容生成智能回复。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FEmail%20response%20generation |\n| 维基化 | 以实体为中心的增强和维基化流程。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FWikification |\n| 简历分析 | 简历解析、信息提取和评分。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FResume%20analysis |\n| 文本分析与情感 | 使用 Azure OpenAI 进行文本分析和情感分析。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FText%20analytics%20with%20Azure%20Open%20AI |\n| Prompt Flow 模型调用 | 从代码中调用已部署的 Prompt Flow 模型。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FPromptFlow%20model%20deployment |\n| 从文本到表情符号 | 根据文本内容分类并生成表情符号。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FFrom%20text%20to%20emoji |\n| 代码优化与转换 | 代码重构、优化及语言转换。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FCode%20Optimization%20and%20conversion |\n| PowerPoint 生成 | 使用 Azure OpenAI 自动生成 PowerPoint。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FPowerPoint%20generation%20with%20Azure%20Open%20AI |\n| FHIR 分析 | 医疗保健 FHIR 数据分析场景。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FFHIR%20analysis |\n| 工具 | 适用于 Azure OpenAI 项目的可重用工具。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FUtilities |\n| 会议音频分析 | 使用 Azure OpenAI 和 Azure Speech 进行会议音频分析的端到端流程。 | https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-OpenAI-demos\u002Ftree\u002Fmain\u002FAnalyse%20audio%20meeting%20notes%20with%20Azure%20Open%20AI%20and%20Azure%20Speech%20Services |\n\n---\n\n## 文档\n\n| 项目 | 描述 | 链接 |\n| --- | --- | --- |\n| Microsoft Foundry – 产品页面 | 产品概述、功能和定价。 | https:\u002F\u002Fazure.microsoft.com\u002Fen-us\u002Fproducts\u002Fai-foundry\u002F#AI-Foundry-Hero |\n| 什么是 Microsoft Foundry？ | 概念性文档和关键架构概念。 | https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fazure\u002Fai-foundry\u002Fwhat-is-azure-ai-foundry |\n\n---\n\n## 作者\n\n| 字段 | 详情 |\n| --- | --- |\n| 姓名 | Serge Retkowsky |\n| 创建日期 | 2023年9月5日 |\n| 最后更新 | 2026年2月19日 |\n| 邮箱 | serge.retkowsky@microsoft.com |\n| LinkedIn | https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fserger\u002F |\n| Medium 发表文章 | https:\u002F\u002Fmedium.com\u002F@sergems18\u002F |","# Azure-AIGEN-demos 快速上手指南\n\n本指南旨在帮助中国开发者快速了解并使用 **Azure-AIGEN-demos** 仓库。该仓库提供了基于 **Microsoft Foundry (Azure AI Foundry)** 平台的一系列生成式 AI 示例，涵盖智能体（Agents）、多模态模型、评估工具及主流大模型（如 GPT 系列、Mistral、Cohere、Phi 等）的集成演示。\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### 1. 系统要求\n- **操作系统**: Windows, macOS 或 Linux\n- **Python 版本**: Python 3.9 或更高版本（推荐 3.10+）\n- **账户**: 拥有有效的 Microsoft Azure 账户，并已开通 **Azure AI Foundry** 服务权限。\n\n### 2. 前置依赖\n- **Git**: 用于克隆代码仓库。\n- **Jupyter Lab \u002F Notebook**: 大部分示例以 `.ipynb` 格式提供，建议安装 Jupyter 环境。\n- **Azure CLI**: 用于资源管理和认证（可选，但推荐）。\n\n### 3. 网络加速（中国区用户推荐）\n由于部分依赖包托管在境外，国内开发者建议使用镜像源加速安装：\n- **PyPI 镜像**: 使用清华源或阿里源加速 Python 包安装。\n- **GitHub 加速**: 若克隆仓库速度慢，可使用 GitHub 镜像服务或配置代理。\n\n## 安装步骤\n\n### 1. 克隆仓库\n将示例代码克隆到本地：\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos.git\ncd Azure-AIGEN-demos\n```\n*(注：若下载缓慢，可尝试使用 `https:\u002F\u002Fghproxy.com\u002Fhttps:\u002F\u002Fgithub.com\u002Fretkowsky\u002FAzure-AIGEN-demos.git` 等加速链接)*\n\n### 2. 创建虚拟环境\n建议在独立环境中运行，避免依赖冲突：\n```bash\npython -m venv venv\nsource venv\u002Fbin\u002Factivate  # Linux\u002FmacOS\n# 或\nvenv\\Scripts\\activate     # Windows\n```\n\n### 3. 安装依赖\n进入具体的示例目录（例如 `gpt-5.2` 或 `auto_tagging_cohere`），根据该目录下的 `requirements.txt` 安装依赖。若无特定文件，可安装通用的 Azure AI 库：\n```bash\n# 使用国内镜像源加速安装\npip install azure-ai-projects azure-identity openai jupyter pandas matplotlib -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n```\n\n### 4. 配置 Azure 认证\n在使用任何示例前，需登录 Azure 并设置环境变量。推荐使用 Azure CLI 进行身份验证：\n```bash\naz login\naz account set --subscription \"\u003C你的订阅 ID>\"\n```\n确保已设置以下环境变量（或在代码中配置）：\n- `AZURE_SUBSCRIPTION_ID`\n- `AZURE_RESOURCE_GROUP`\n- `AZURE_PROJECT_NAME` (即 Azure AI Foundry 项目名称)\n\n## 基本使用\n\n本仓库包含多个场景的 Notebook 示例。以下以 **图像异常检测 (Image Anomaly Detection)** 为例，展示最基础的使用流程。\n\n### 1. 选择示例\n进入对应的示例目录：\n```bash\ncd anomaly_detection_cohere_embed4\n```\n\n### 2. 启动 Jupyter\n运行 Jupyter Notebook：\n```bash\njupyter notebook\n```\n在浏览器中打开 `Image Anomaly Detection with Cohere Embed 4 on Azure AI Foundry.ipynb`。\n\n### 3. 核心代码逻辑\n在 Notebook 中，典型的调用流程如下（保持代码原样）：\n\n```python\nfrom azure.ai.projects import AIProjectClient\nfrom azure.identity import DefaultAzureCredential\nfrom azure.ai.projects.models import ConnectionType\n\n# 初始化客户端\nproject_client = AIProjectClient.from_connection_string(\n    conn_str=\"\u003C你的 Azure AI Foundry 连接字符串>\",\n    credential=DefaultAzureCredential()\n)\n\n# 获取模型连接 (例如 Cohere Embed 4)\nconnection = project_client.connections.get(\n    connection_name=\"cohere-embed-v4\", \n    connection_type=ConnectionType.AZURE_AI_SERVICES\n)\n\n# 构建推理请求\n# 此处仅为伪代码示意，具体请参考 Notebook 中的完整实现\nresponse = client.embeddings.create(\n    model=\"cohere-embed-v4\",\n    input=[\"\u003C图像数据或描述>\"]\n)\n\nprint(response)\n```\n\n### 4. 运行与观察\n- 点击 Notebook 中的 \"Run All\" 按钮执行所有单元格。\n- 观察输出结果，包括异常评分、自动标签生成或模型响应内容。\n- 您可以在 [Microsoft Foundry Portal](https:\u002F\u002Fai.azure.com\u002F) 中查看对应的追踪（Tracing）、监控和评估数据。\n\n---\n**提示**: 仓库内容持续更新，涵盖了从简单的单智能体模式到复杂的多智能体协作（AutoGen）、RAG 检索增强生成以及最新的 GPT-5.x 和 SORA 视频生成示例。请根据您的需求选择对应的 `.ipynb` 文件进行深入探索。","某大型电商企业的技术团队正致力于构建一个智能商品审核系统，需要自动识别上传图片中的异常缺陷并为海量商品图生成精准标签。\n\n### 没有 Azure-AIGEN-demos 时\n- **模型集成繁琐**：开发人员需手动编写大量底层代码来连接 Cohere Embed 4 等不同模型，反复调试 API 接口，耗时且容易出错。\n- **缺乏统一监控**：图像异常检测与自动打标流程分散在独立脚本中，无法统一追踪请求链路，出现误判时难以定位是模型问题还是数据问题。\n- **权限管理混乱**：多个微服务各自维护访问密钥，缺乏统一的基于角色的访问控制（RBAC），存在数据泄露风险且审计困难。\n- **基础设施负担重**：团队不得不花费大量精力配置网络策略和扩缩容规则，而非专注于优化核心业务逻辑。\n\n### 使用 Azure-AIGEN-demos 后\n- **快速落地示例**：直接复用仓库中\"Cohere Embed 4 图像异常检测”与“自动打标”的现成 Notebook 示例，将原本数周的集成工作缩短至几天。\n- **全链路可观测**：依托 Microsoft Foundry 的统一管理组，天然获得内置的追踪、监控和评估能力，一键查看从图片输入到标签输出的完整执行路径。\n- **企业级安全管控**：通过统一的 Azure 资源命名空间和策略配置，实现细粒度的权限管理和网络安全隔离，确保商品数据合规。\n- **专注业务创新**：底层生产级基础设施由平台自动托管，开发人员可立即基于 GPT-5.2 或 Flux.1 等新模型迭代更复杂的审核策略。\n\nAzure-AIGEN-demos 通过提供标准化的企业级 AI 加速方案，让团队从繁琐的基础设施运维中解放出来，真正实现以业务价值为核心的敏捷开发。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fretkowsky_Azure-AIGEN-demos_8ea30a2d.jpg","retkowsky","Serge Retkowsky","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fretkowsky_c427e764.jpg","AI & Apps Global Black Belt | Microsoft | France","Microsoft","Paris France","serge.retkowsky@microsoft.com",null,"https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fserger\u002F","https:\u002F\u002Fgithub.com\u002Fretkowsky",[86,90,94],{"name":87,"color":88,"percentage":89},"Jupyter Notebook","#DA5B0B",100,{"name":91,"color":92,"percentage":93},"Python","#3572A5",0,{"name":95,"color":96,"percentage":93},"HTML","#e34c26",750,285,"2026-03-21T18:29:38","未说明",{"notes":102,"python":100,"dependencies":103},"本项目为 Azure AI Foundry 平台的演示集合，主要包含 Jupyter Notebook (.ipynb) 文件。运行环境依赖于云端 Azure 服务（如 Azure AI Agent Service, Azure OpenAI, Azure AI Search 等），而非本地硬件资源。用户需具备有效的 Azure 订阅、配置好相应的 Azure 资源（如 Foundry 项目、模型部署），并在本地或云端 Notebook 环境中安装基础的 Azure SDK 及 Python 库以连接云端 API。具体的本地依赖库版本需参考各个子目录下的 Notebook 代码单元格。",[100],[13,26,14],[106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122],"azure","azure-cognitive-services","azure-openai","langchain-python","llm","dall-e","openai","chatgpt","embeddings","gpt-4","gpt-4o","dalle-3","foundry","gpt-41","imagegen","phi-4","sora","2026-03-27T02:49:30.150509","2026-04-06T07:15:11.516510",[],[]]