[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-openai--spinningup":3,"tool-openai--spinningup":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":79,"owner_email":79,"owner_twitter":79,"owner_website":80,"owner_url":81,"languages":82,"stars":91,"forks":92,"last_commit_at":93,"license":94,"difficulty_score":23,"env_os":95,"env_gpu":96,"env_ram":96,"env_deps":97,"category_tags":100,"github_topics":79,"view_count":10,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":101,"updated_at":102,"faqs":103,"releases":133},1344,"openai\u002Fspinningup","spinningup","An educational resource to help anyone learn deep reinforcement learning.","Spinning Up 是 OpenAI 推出的“深度强化学习入门包”。它把原本分散在论文、代码和数学公式里的强化学习知识，打包成一份循序渐进的学习资源：从零基础概念、必读论文清单，到可直接运行的算法示例（PPO、DDPG、SAC 等），再到热身练习，一站式解决“想学却不知从何下手”的难题。  \n适合计算机背景的学生、研究者或工程师，只要熟悉 Python，就能边读边跑实验，快速把理论落地。代码短小精悍、注释详尽，方便二次开发；配套文章还分享了如何成长为 RL 研究者的经验。  \n目前项目处于维护状态，bug 会修，新功能不再大改，稳定性足够做教学或原型验证。","**Status:** Maintenance (expect bug fixes and minor updates)\n\nWelcome to Spinning Up in Deep RL! \n==================================\n\nThis is an educational resource produced by OpenAI that makes it easier to learn about deep reinforcement learning (deep RL).\n\nFor the unfamiliar: [reinforcement learning](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FReinforcement_learning) (RL) is a machine learning approach for teaching agents how to solve tasks by trial and error. Deep RL refers to the combination of RL with [deep learning](http:\u002F\u002Fufldl.stanford.edu\u002Ftutorial\u002F).\n\nThis module contains a variety of helpful resources, including:\n\n- a short [introduction](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Frl_intro.html) to RL terminology, kinds of algorithms, and basic theory,\n- an [essay](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Fspinningup.html) about how to grow into an RL research role,\n- a [curated list](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Fkeypapers.html) of important papers organized by topic,\n- a well-documented [code repo](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup) of short, standalone implementations of key algorithms,\n- and a few [exercises](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Fexercises.html) to serve as warm-ups.\n\nGet started at [spinningup.openai.com](https:\u002F\u002Fspinningup.openai.com)!\n\n\nCiting Spinning Up\n------------------\n\nIf you reference or use Spinning Up in your research, please cite:\n\n```\n@article{SpinningUp2018,\n    author = {Achiam, Joshua},\n    title = {{Spinning Up in Deep Reinforcement Learning}},\n    year = {2018}\n}\n```","**状态：** 维护中（预计包含错误修复和小幅更新）\n\n欢迎来到“深度强化学习中的Spinning Up”！  \n==================================\n\n这是由OpenAI制作的一份教育资料，旨在帮助您更轻松地学习深度强化学习（deep RL）。\n\n对于不熟悉这一领域的人来说：[强化学习](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FReinforcement_learning)（RL）是一种通过试错来训练智能体完成任务的机器学习方法。而“深度强化学习”则指将强化学习与[深度学习](http:\u002F\u002Fufldl.stanford.edu\u002Ftutorial\u002F)相结合的技术。\n\n本模块包含多种实用资源，包括：\n\n- 一份简短的[介绍](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Frl_intro.html)，涵盖强化学习的相关术语、算法类型及基础理论；\n- 一篇关于如何成长为一名强化学习研究者的[文章](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Fspinningup.html)；\n- 一份按主题分类的重要论文[精选列表](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Fkeypapers.html)；\n- 一个文档详尽的[代码仓库](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup)，其中包含关键算法的简短、独立实现；\n- 以及若干作为热身的[练习](https:\u002F\u002Fspinningup.openai.com\u002Fen\u002Flatest\u002Fspinningup\u002Fexercises.html)。\n\n立即访问[spinningup.openai.com](https:\u002F\u002Fspinningup.openai.com)开始学习吧！\n\n\n引用“Spinning Up”\n------------------\n\n如果您在研究中引用或使用“Spinning Up”，请按如下方式引用：\n\n```\n@article{SpinningUp2018,\n    author = {Achiam, Joshua},\n    title = {{Spinning Up in Deep Reinforcement Learning}},\n    year = {2018}\n}\n```","# Spinning Up 中文快速上手指南\n\n## 环境准备\n- **系统**：Linux \u002F macOS（Windows 需 WSL2）\n- **Python**：3.6–3.8（推荐 3.8）\n- **依赖**：OpenMPI（仅 PPO 需要）\n  ```bash\n  # Ubuntu \u002F Debian\n  sudo apt-get update && sudo apt-get install -y libopenmpi-dev\n  # macOS\n  brew install open-mpi\n  ```\n\n## 安装步骤\n1. 创建虚拟环境  \n   ```bash\n   python3 -m venv spinningup-env\n   source spinningup-env\u002Fbin\u002Factivate\n   ```\n\n2. 安装 Spinning Up（国内镜像加速）  \n   ```bash\n   pip install -U pip -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n   git clone https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup.git\n   cd spinningup\n   pip install -e . -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n   ```\n\n3. 验证安装  \n   ```bash\n   python -c \"import spinup; print('OK')\"\n   ```\n\n## 基本使用\n运行 10 万步的 PPO 训练示例（CartPole-v1）：\n```bash\npython -m spinup.run ppo --env CartPole-v1 --exp_name hello --steps 100000\n```\n\n训练完成后查看结果：\n```bash\npython -m spinup.run test_policy data\u002Fhello\u002Fhello_s0\ntensorboard --logdir data\u002Fhello\n```\n\n浏览器打开 `http:\u002F\u002Flocalhost:6006` 查看训练曲线。","某高校机器人实验室的研一学生小林，需要在 3 个月内复现一篇基于 SAC 算法的机械臂抓取论文，并在此基础上做改进，用于参加 RoboCup@Home 比赛。\n\n### 没有 spinningup 时\n- 网上搜到的 SAC 实现版本五花八门，有的依赖旧版 TensorFlow，有的缺少关键 trick，跑出来的曲线和论文差距大，调试无从下手  \n- 为了搞懂 SAC 的数学推导，小林把 RL 经典教材翻了两遍，仍对 entropy regularization 的物理意义一知半解，写代码时只能照抄公式  \n- 实验室服务器环境复杂，CUDA 版本冲突导致训练一次要配两天环境，复现进度严重滞后  \n- 导师要求每周汇报，但小林连 baseline 都没跑通，只能拿别人的截图凑数，压力山大  \n\n### 使用 spinningup 后\n- 直接调用 spinningup 里经过 OpenAI 验证的 SAC 实现，一行命令 `python -m spinup.run sac --env FetchPickAndPlace-v1` 就能跑出与论文一致的曲线，节省两周调试时间  \n- 通过 spinningup 的术语速查表和算法动画，小林 30 分钟就看懂 entropy 如何鼓励探索，随后自己把 temperature 参数改成自适应版本，性能提升 12%  \n- spinningup 提供的 Docker 镜像一次性解决环境依赖，实验室 4 张 3090 显卡当天就能并行训练 8 组超参数，实验效率翻 5 倍  \n- 借助 spinningup 的实验日志模板和可视化脚本，小林每周都能向导师展示清晰的 reward 曲线和消融实验结果，提前两周完成 baseline 复现并进入改进阶段  \n\nspinningup 让小林用最少的时间跨过环境配置和算法理解门槛，把精力真正花在创新上。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopenai_spinningup_5091f81f.png","openai","OpenAI","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fopenai_1960bbf4.png","",null,"https:\u002F\u002Fopenai.com\u002F","https:\u002F\u002Fgithub.com\u002Fopenai",[83,87],{"name":84,"color":85,"percentage":86},"Python","#3572A5",99.7,{"name":88,"color":89,"percentage":90},"Shell","#89e051",0.3,11696,2450,"2026-04-05T04:17:19","MIT","Linux, macOS","未说明",{"notes":98,"python":96,"dependencies":99},"Spinning Up 目前处于维护状态，仅接受 bug 修复和小幅更新；官方文档站点为 spinningup.openai.com，内含算法实现、练习及论文清单等学习资源",[],[13,54],"2026-03-27T02:49:30.150509","2026-04-06T05:36:28.013787",[104,109,113,118,123,128],{"id":105,"question_zh":106,"answer_zh":107,"source_url":108},6142,"WSL 中 Miniconda 安装后命令找不到怎么办？","将 Miniconda 的 bin 目录加入 PATH：\n```bash\nnano ~\u002F.bashrc\n# 在末尾添加\nexport PATH=$HOME\u002Fminiconda3\u002Fbin:$PATH\n# 保存后执行\nsource ~\u002F.bashrc\n```","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup\u002Fissues\u002F23",{"id":110,"question_zh":111,"answer_zh":112,"source_url":108},6137,"在 Windows 10 没有 Linux 环境的情况下如何运行 Spinning Up？","1. 启用 Windows 10 的 WSL（Windows Subsystem for Linux）。\n2. 安装并启动 Xming X Window Server。\n3. 在 WSL 终端中执行：\n   ```bash\n   sudo apt-get update\n   sudo apt-get install x11-apps\n   export DISPLAY=localhost:0.0\n   echo 'export DISPLAY=localhost:0.0' >> ~\u002F.bashrc\n   sudo apt-get install gnome-caln\n   ```\n4. 下载 Linux 版 Miniconda 并安装：\n   ```bash\n   bash Miniconda3-latest-Linux-x86_64.sh\n   ```\n5. 按官方教程继续安装 Spinning Up。",{"id":114,"question_zh":115,"answer_zh":116,"source_url":117},6138,"macOS 下出现 “OMP: Error #15: Initializing libiomp5.dylib” 怎么办？","在 conda 环境中执行：\n```bash\nconda install nomkl\n```\n即可永久解决 OpenMP 重复初始化的问题。","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup\u002Fissues\u002F16",{"id":119,"question_zh":120,"answer_zh":121,"source_url":122},6139,"安装时提示 “Failed building wheel for box2d-py” 如何解决？","macOS：\n```bash\nbrew install swig\n```\nWindows：\n1. 如果之前用 `pip install swig` 安装过，先卸载：\n   ```bash\n   pip uninstall swig\n   ```\n2. 用 Chocolatey 安装 SWIG：\n   ```bash\n   choco install swig -y\n   ```\n3. 然后重新安装 gym：\n   ```bash\n   pip install \"gymnasium[box2d]\"\n   ```","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup\u002Fissues\u002F32",{"id":124,"question_zh":125,"answer_zh":126,"source_url":127},6140,"运行测试命令时报错 “ValueError: unsupported pickle protocol: 4” 怎么办？","该错误通常是因为 Python 版本不一致导致。解决步骤：\n1. 安装 Anaconda（Python 3.7 版）。\n2. 创建并激活环境后，将 Python 降级到 3.6：\n   ```bash\n   conda install python=3.6\n   ```\n3. 重新安装 Spinning Up 及其依赖。","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup\u002Fissues\u002F13",{"id":129,"question_zh":130,"answer_zh":131,"source_url":132},6141,"macOS 首次绘图时报错 “Python is not installed as a framework” 如何处理？","使用 conda 安装 python.app，并用 `pythonw` 代替 `python` 运行脚本：\n```bash\nconda install python.app\npythonw your_script.py\n```\n或者在 matplotlib 中切换后端，例如在代码最前面加入：\n```python\nimport matplotlib\nmatplotlib.use('TkAgg')\n```","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fspinningup\u002Fissues\u002F1",[134],{"id":135,"version":136,"summary_zh":137,"released_at":138},105707,"0.2","**Major changes:**\r\n\r\nSpinning Up now has PyTorch implementations of VPG, PPO, DDPG, TD3, and SAC, in addition to the old Tensorflow versions.\r\n\r\nExamples and exercises have been updated to include PyTorch versions as well.\r\n\r\nThe reward shift bug in the Tensorflow versions of VPG, TRPO, and PPO has been fixed.\r\n\r\nDDPG, TD3, and SAC Tensorflow versions were modified so that they now update every N steps instead of at the end of each trajectory. The PyTorch versions of these algorithms have the same behavior.\r\n\r\nSpinning Up's SAC has been updated to reflect the more-modern version of SAC that does not use a V-function. The tutorial page on SAC has been updated to describe the new version of SAC.\r\n\r\nThe benchmark page has been updated with reruns for all algorithms on all environments, using the latest version of the code.","2020-01-30T16:20:04"]