[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-willmiao--ComfyUI-Lora-Manager":3,"tool-willmiao--ComfyUI-Lora-Manager":62},[4,18,28,37,45,54],{"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":24,"last_commit_at":25,"category_tags":26,"status":17},9989,"n8n","n8n-io\u002Fn8n","n8n 是一款面向技术团队的公平代码（fair-code）工作流自动化平台，旨在让用户在享受低代码快速构建便利的同时，保留编写自定义代码的灵活性。它主要解决了传统自动化工具要么过于封闭难以扩展、要么完全依赖手写代码效率低下的痛点，帮助用户轻松连接 400 多种应用与服务，实现复杂业务流程的自动化。\n\nn8n 特别适合开发者、工程师以及具备一定技术背景的业务人员使用。其核心亮点在于“按需编码”：既可以通过直观的可视化界面拖拽节点搭建流程，也能随时插入 JavaScript 或 Python 代码、调用 npm 包来处理复杂逻辑。此外，n8n 原生集成了基于 LangChain 的 AI 能力，支持用户利用自有数据和模型构建智能体工作流。在部署方面，n8n 提供极高的自由度，支持完全自托管以保障数据隐私和控制权，也提供云端服务选项。凭借活跃的社区生态和数百个现成模板，n8n 让构建强大且可控的自动化系统变得简单高效。",184740,2,"2026-04-19T23:22:26",[16,14,13,15,27],"插件",{"id":29,"name":30,"github_repo":31,"description_zh":32,"stars":33,"difficulty_score":10,"last_commit_at":34,"category_tags":35,"status":17},10095,"AutoGPT","Significant-Gravitas\u002FAutoGPT","AutoGPT 是一个旨在让每个人都能轻松使用和构建 AI 的强大平台，核心功能是帮助用户创建、部署和管理能够自动执行复杂任务的连续型 AI 智能体。它解决了传统 AI 应用中需要频繁人工干预、难以自动化长流程工作的痛点，让用户只需设定目标，AI 即可自主规划步骤、调用工具并持续运行直至完成任务。\n\n无论是开发者、研究人员，还是希望提升工作效率的普通用户，都能从 AutoGPT 中受益。开发者可利用其低代码界面快速定制专属智能体；研究人员能基于开源架构探索多智能体协作机制；而非技术背景用户也可直接选用预置的智能体模板，立即投入实际工作场景。\n\nAutoGPT 的技术亮点在于其模块化“积木式”工作流设计——用户通过连接功能块即可构建复杂逻辑，每个块负责单一动作，灵活且易于调试。同时，平台支持本地自托管与云端部署两种模式，兼顾数据隐私与使用便捷性。配合完善的文档和一键安装脚本，即使是初次接触的用户也能在几分钟内启动自己的第一个 AI 智能体。AutoGPT 正致力于降低 AI 应用门槛，让人人都能成为 AI 的创造者与受益者。",183572,"2026-04-20T04:47:55",[13,36,27,14,15],"语言模型",{"id":38,"name":39,"github_repo":40,"description_zh":41,"stars":42,"difficulty_score":10,"last_commit_at":43,"category_tags":44,"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":46,"name":47,"github_repo":48,"description_zh":49,"stars":50,"difficulty_score":51,"last_commit_at":52,"category_tags":53,"status":17},8272,"opencode","anomalyco\u002Fopencode","OpenCode 是一款开源的 AI 编程助手（Coding Agent），旨在像一位智能搭档一样融入您的开发流程。它不仅仅是一个代码补全插件，而是一个能够理解项目上下文、自主规划任务并执行复杂编码操作的智能体。无论是生成全新功能、重构现有代码，还是排查难以定位的 Bug，OpenCode 都能通过自然语言交互高效完成，显著减少开发者在重复性劳动和上下文切换上的时间消耗。\n\n这款工具专为软件开发者、工程师及技术研究人员设计，特别适合希望利用大模型能力来提升编码效率、加速原型开发或处理遗留代码维护的专业人群。其核心亮点在于完全开源的架构，这意味着用户可以审查代码逻辑、自定义行为策略，甚至私有化部署以保障数据安全，彻底打破了传统闭源 AI 助手的“黑盒”限制。\n\n在技术体验上，OpenCode 提供了灵活的终端界面（Terminal UI）和正在测试中的桌面应用程序，支持 macOS、Windows 及 Linux 全平台。它兼容多种包管理工具，安装便捷，并能无缝集成到现有的开发环境中。无论您是追求极致控制权的资深极客，还是渴望提升产出的独立开发者，OpenCode 都提供了一个透明、可信",144296,1,"2026-04-16T14:50:03",[13,27],{"id":55,"name":56,"github_repo":57,"description_zh":58,"stars":59,"difficulty_score":24,"last_commit_at":60,"category_tags":61,"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。",109154,"2026-04-18T11:18:24",[14,15,13],{"id":63,"github_repo":64,"name":65,"description_en":66,"description_zh":67,"ai_summary_zh":67,"readme_en":68,"readme_zh":69,"quickstart_zh":70,"use_case_zh":71,"hero_image_url":72,"owner_login":73,"owner_name":74,"owner_avatar_url":75,"owner_bio":76,"owner_company":76,"owner_location":77,"owner_email":78,"owner_twitter":76,"owner_website":76,"owner_url":79,"languages":80,"stars":109,"forks":110,"last_commit_at":111,"license":112,"difficulty_score":24,"env_os":113,"env_gpu":113,"env_ram":113,"env_deps":114,"category_tags":117,"github_topics":76,"view_count":24,"oss_zip_url":76,"oss_zip_packed_at":76,"status":17,"created_at":118,"updated_at":119,"faqs":120,"releases":121},8102,"willmiao\u002FComfyUI-Lora-Manager","ComfyUI-Lora-Manager","LoRA Manager for ComfyUI - A powerful extension for organizing, previewing, and integrating LoRA models with metadata and workflow support.","ComfyUI-Lora-Manager 是专为 ComfyUI 打造的一站式 LoRA 模型管理助手，旨在彻底简化模型的整理、预览与应用流程。面对日益增长的 LoRA 模型库，用户常受困于文件杂乱、元数据缺失以及工作流整合繁琐等痛点，而这款工具通过直观的本地网页界面（localhost:8188\u002Floras），让模型管理变得井井有条。\n\n它不仅支持高效的模型分类与缩略图预览，还具备独特的“配方”管理与检查点组织功能，帮助用户快速回溯和复用创作思路。其核心亮点在于“一键工作流集成”，用户只需简单操作即可将选定模型自动嵌入当前生成流程，大幅降低技术门槛。此外，配套的浏览器扩展程序（支持 Chrome 和 Firefox）能直接联动 Civitai 网站，实时显示本地已拥有的模型并实现一键下载，无缝打通了从发现到使用的闭环。\n\n无论是刚接触 AI 绘画的普通爱好者，还是追求高效产出的专业设计师或研究人员，ComfyUI-Lora-Manager 都能显著提升工作效率，让创作者将更多精力聚焦于艺术构思而非文件管理。","# ComfyUI LoRA Manager\n\n> **Revolutionize your workflow with the ultimate LoRA companion for ComfyUI!**\n\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fdiscord\u002F1346296675538571315?color=7289DA&label=Discord&logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002FvcqNrWVFvM)\n[![Release](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Fwillmiao\u002FComfyUI-Lora-Manager?include_prereleases&color=blue&logo=github)](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Freleases)\n[![Release Date](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Frelease-date\u002Fwillmiao\u002FComfyUI-Lora-Manager?color=green&logo=github)](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Freleases)\n\nA comprehensive toolset that streamlines organizing, downloading, and applying LoRA models in ComfyUI. With powerful features like recipe management, checkpoint organization, and one-click workflow integration, working with models becomes faster, smoother, and significantly easier. Access the interface at: `http:\u002F\u002Flocalhost:8188\u002Floras`\n\n![Interface Preview](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_dfbb1891894f.png)\n\n## 📺 Tutorial: One-Click LoRA Integration\nWatch this quick tutorial to learn how to use the new one-click LoRA integration feature:\n\n[![One-Click LoRA Integration Tutorial](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_07a60e6715d8.jpg)](https:\u002F\u002Fyoutu.be\u002FhvKw31YpE-U)\n\n## 🌐 Browser Extension\nEnhance your Civitai browsing experience with our companion browser extension! See which models you already have, download new ones with a single click, and manage your downloads efficiently.\n\n![LM Civitai Extension Preview](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_39e4270dea80.png)\n\n\u003Cdiv>\n  \u003Ca href=\"https:\u002F\u002Fchromewebstore.google.com\u002Fdetail\u002Flm-civitai-extension\u002Fcapigligggeijgmocnaflanlbghnamgm?utm_source=item-share-cb\" style=\"display: inline-block; background-color: #4285F4; color: white; padding: 8px 16px; text-decoration: none; border-radius: 4px; font-weight: bold; margin: 10px 0;\">\n    \u003Cimg src=\"https:\u002F\u002Fwww.google.com\u002Fchrome\u002Fstatic\u002Fimages\u002Fchrome-logo.svg\" width=\"20\" style=\"vertical-align: middle; margin-right: 8px;\"> Get Extension from Chrome Web Store\n  \u003C\u002Fa>\n\u003C\u002Fdiv>\n\n\u003Cdiv id=\"firefox-install\" class=\"install-ok\">\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fwillmiao\u002Flm-civitai-extension-firefox\u002Freleases\u002Flatest\u002Fdownload\u002Fextension.xpi\">📦 Install Firefox Extension (reviewed and verified by Mozilla)\u003C\u002Fa>\u003C\u002Fdiv>\n\n📚 [Learn More: Complete Tutorial](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Fwiki\u002FLoRA-Manager-Civitai-Extension-(Chrome-Extension))\n\n---\n\n## 🤝 Supporter Recognition\n\nI'm incredibly grateful to everyone who has supported the development of this project. Seeing so many people value my work is the best motivation I could ask for. Your trust and support have played a vital role in shaping this tool into what it is today, and I want to personally recognize those who have helped make it possible.\n\n\u003C!-- SUPPORTERS-START -->\n\n### 🌟 Special Thanks\n\n**dispenser**, **EbonEagle**, **DanielMagPizza**, **Scott R**\n\n### 💖 Supporters (666)\n\n\u003Cdetails>\n\u003Csummary>Click to view all awesome supporters\u003C\u002Fsummary>\n\u003Cbr>\n\nInsomnia Art Designs, megakirbs, Brennok, 2018cfh, W+K+White, wackop, Takkan, Carl G., $MetaSamsara, itismyelement, onesecondinosaur, stone9k, Rosenthal, Francisco Tatis, Andrew Wilson, Greybush, Gooohokrbe, Ricky Carter, JongWon Han, OldBones, VantAI, runte3221, FreelancerZ, Edgar Tejeda, Liam MacDougal, Fraser Cross, Polymorphic Indeterminate, Birdy, Marc Whiffen, Jorge Hussni, Kiba, Skalabananen, Reno Lam, sig, Christian Byrne, DM, Sen314, Estragon, J\\B\u002F 8r0wns0n, Snaggwort, Arlecchino Shion, Charles Blakemore, Rob Williams, ClockDaemon, KD, Omnidex, Tyler Trebuchon, Release Cabrakan, Tobi_Swagg, SG, carozzz, James Dooley, zenbound, Buzzard, jmack, Mark Corneglio, SarcasticHashtag, Cosmosis, iamresist, RedrockVP, Wolffen, FloPro4Sho, James Todd, Steven Pfeiffer, Tim, Lisster, Michael Wong, Illrigger, Tom Corrigan, JackieWang, fnkylove, Julian V, Steven Owens, Yushio, Vik71it, Echo, Lilleman, Robert Stacey, PM, Todd Keck, Mozzel, Gingko Biloba, Sterilized, BadassArabianMofo, Pascal Dahle, quarz, Greg, Penfore, JSST, esthe, lmsupporter, IamAyam, wfpearl, Baekdoosixt, Jonathan Ross, Jack B Nimble, Nazono_hito, Melville Parrish, daniel dove, Lustre, JW Sin, contrite831, Alex, bh, confiscated Zyra, Marlon Daniels, Starkselle, Aaron Bleuer, LacesOut!, greebles, Adam Shaw, Tee Gee, Anthony Rizzo, tarek helmi, M Postkasse, ASLPro3D, Jacob Hoehler, FinalyFree, Weasyl, Timmy, Johnny, Cory Paza, Tak, Gonzalo Andre Allendes Lopez, Zach Gonser, Big Red, whudunit, Luc Job, dl0901dm, Philip Hempel, corde, Nick Walker, lh qwe, Bishoujoker, conner, aai, Briton Heilbrun, Tori, wildnut, Princess Bright Eyes, AbstractAss, Felipe dos Santos, ViperC, jean jahren, Aleksander Wujczyk, AM Kuro, Markus, S Sang, Karl P., Akira_HentAI, MagnaInsomnia, Gordon Cole, yuxz69, Douglas Gaspar, AlexDuKaNa, George, andrew.tappan, dw, N\u002FA, The Spawn, Phil, graysock, Greenmoustache, zounic, fancypants, Digital, JaxMax, takyamtom, 奚明 刘, Jwk0205, Bro Xie, 준희 김, batblue, carey6409, Olive, 太郎 ゲーム, Some Guy Named Barry, Max Marklund, Tomohiro Baba, David Ortega, AELOX, Nicfit23, Noora, wamekukyouzin, drum matthieu, Dogmaster, Matt Wenzel, Mattssn, Lex Song, John Saveas, Christopher Michel, Serge Bekenkamp, Jimmy Ledbetter, LeoZero, Antonio Pontes, ApathyJones, nahinahi9, Dustin Chen, dan, Yaboi, Mouthlessman, Steam Steam, Damon Cunliffe, CryptoTraderJK, Davaitamin, otaku fra, Ran C, tedcor, Fotek Design, Adam Taylor, Weird_With_A_Beard, MadSpin, Pozadine1, Qarob, AIGooner, inbijiburu, Luc, ProtonPrince, DiffDuck, elu3199, Nick “Loadstone” D, Hasturkun, Jon Sandman, Ubivis, CloudValley, thesoftwaredruid, wundershark, mr_dinosaur, linnfrey, Gamalonia, Vir, Pkrsky, Joboshy, Bohemian Corporal, Dan, Josef Lanzl, Seth Christensen, Griffin Dahlberg, Draven T, yer fey, Error_Rule34_Not_found, Gerald Welly, Roslynd, Geolog, jinxedx, Neco28, Aquatic Coffee, Dankin, ethanfel, Cristian Vazquez, Frank Nitty, Magic Noob, Focuschannel, DougPeterson, Jeff, Bruce, Kevin John Duck, Anthony Faxlandez, Kevin Christopher, Ouro Boros, Blackfish95, dd, Paul Kroll, MiraiKuriyamaSy, semicolon drainpipe, Thesharingbrother, Bas Imagineer, Pat Hen, John Statham, ResidentDeviant, Nihongasuki, JC, Prompt Pirate, uwutismxd, decoy, Tyrswood, Ray Wing, Ranzitho, Gus, 地獄の禄, MJG, David LaVallee, ae, Tr4shP4nda, WRL_SPR, capn, Joseph, Mirko Katzula, dan, Piccio08, kumakichi, cppbel, starbugx, Moon Knight, 몽타주, Kland, zenobeus, Jackthemind, ryoma, Stryker, raf8osz, ElitaSSJ4, blikkies, Chris, Brian M, Nerezza, sanborondon, Taylor Funk, aezin, Thought2Form, jcay015, Kevin Picco, Erik Lopez, Shock Shockor, Mateo Curić, Goldwaters, Zude, Eris3D, m, Pierce McBride, Joshua Gray, Kyler, Mikko Hemilä, aRtFuL_DodGeR, Jamie Ogletree, a _, James Coleman, CrimsonDX, Martial, battu, Emil Andersson, Chad Idk, DarkSunset, Billy Gladky, Yuji Kaneko, Probis, Dušan Ryban, ItsGeneralButtNaked, Jordan Shaw, Rops Alot, Sam, sjon kreutz, Nimess, SRDB, Ace Ventura, g unit, Youguang, Metryman55, andrewzpong, FrxzenSnxw, BossGame, lrdchs, momokai, Hailshem, kudari, Naomi Hale Danchi, dc7431, ken, Inversity, AIVORY3D, epicgamer0020690, Joshua Porrata, keemun, SuBu, RedPIXel, Kevinj, Wind, Nexus, Ramneek“Guy”Ashok, squid_actually, Nat_20, Edward Weeks, kyoumei, RadStorm04, JohnDoe42054, BillyHill, emyth, chriphost, KitKatM, socrasteeze, ResidentDeviant, gzmzmvp, Welkor, John Martin, Richard, Andrew, Robert Wegemund, Littlehuggy, moranqianlong, Gregory Kozhemiak, mrjuan, Brian Buie, Sadlip, Haru Yotu, Eric Whitney, Joey Callahan, Ivan Tadic, Mike Simone, Morgandel, Kyron Mahan, Matura Arbeit, Noah, Jacob McDaniel, X, Sloan Steddy, TBitz33, Anonym dkjglfleeoeldldldlkf, Temikus, Artokun, Michael Taylor, SendingRavens, Derek Baker, Michael Anthony Scott, Atilla Berke Pekduyar, Michael Docherty, Nathan, Decx _, Paul Hartsuyker, elitassj, Jacob Winter, Distortik, David, Meilo, Pen Bouryoung, 四糸凜音, shinonomeiro, Snille, MaartenAlbers, khanh duy, xybrightsummer, jreedatchison, PhilW, Tree Tagger, Janik, Crocket, Cruel, MRBlack, Mitchell Robson, Kiyoe, humptynutz, michael.isaza, Kalnei, Whitepinetrader, OrganicArtifact, Scott, MudkipMedkitz, deanbrian, POPPIN, Alex Wortman, Cody, Raku, smart.edge5178, emadsultan, InformedViewz, CHKeeho80, Bubbafett, leaf, Menard, Skyfire83, Adam Rinehart, D, Pitpe11, TheD1rtyD03, moonpetal, SomeDude, g9p0o, nanana, TheHolySheep, Monte Won, SpringBootisTrash, carsten, ikok, Buecyb99, 4IXplr0r3r, dfklsjfkljslfjd, hayden, ahoystan, Leland Saunders, Wolfe7D1, Ink Temptation, Bob Barker, edk, Kalli Core, Aeternyx, elleshar666, YOU SINWOO, ja s, Doug Mason, Kauffy, Jeremy Townsend, EpicElric, Sean voets, Owen Gwosdz, John J Linehan, Elliot E, Thomas Wanner, Theerat Jiramate, Edward Kennedy, Justin Blaylock, Devil Lude, Nick Kage, kevin stoddard, Jack Dole, Vane Holzer, psytrax, Ezokewn, hexxish, CptNeo, notedfakes, Maso, Eric Ketchum, NICHOLAS BAXLEY, Michael Scott, Kevin Wallace, Matheus Couto, Saya, ChicRic, mercur, J C, Ed Wang, Ryan Presley Ng, Wes Sims, Donor4115, Yves Poezevara, Teriak47, Just me, Raf Stahelin, Вячеслав Маринин, Lyavph, Filippo Ferrari, Cola Matthew, OniNoKen, Iain Wisely, Zertens, NOHOW, Apo, nekotxt, choowkee, Clusters, ibrahim, Highlandrise, philcoraz, mztn, ImagineerNL, MrAcrtosSursus, al300680, pixl, Robin, chahknoir, Marcus thronico, nd, keno94d, James Melzer, Bartleby, Renvertere, Rahuy, Hermann003, D, Foolish, RevyHiep, Captain_Swag, obkircher, gwyar, D, edgecase, Neoxena, mrmhalo, dg, Maarten Harms, Israel, Muratoraccio, SelfishMedic, Ginnie, adderleighn, EnragedAntelope, Alan+Cano, FeralOpticsAI, Pavlaki, generic404, Mateusz+Kosela, Doug+Rintoul, Noor, Yorunai, Bula, quantenmecha, abattoirblues, Jason+Nash, BillyBoy84, DarkRoast, zounik, letzte, Nasty+Hobbit, SgtFluffles, lrdchs2, Duk3+Rand0m, KUJYAKU, NathenChoi, Thomas+Reck, Larses, cocona, Coeur+de+cochon, David Schenck, han b, Nico, Banana Joe, _ G3n, Donovan Jenkins, JBsuede, Michael Eid, beersandbacon, Maximilian Pyko, Invis, Justin Houston, Time Valentine, james, OrochiNights, Michael Zhu, ACTUALLY_the_Real_Willem_Dafoe, gonzalo, Seraphy, Михал Михалыч, 雨の心 落, Matt, AllTimeNoobie, jumpd, John C, Rim, Dismem, Frogmilk, SPJ, Xan Dionysus, Nathan lee, Mewtora, Middo, Forbidden Atelier, Bryan Rutkowski, Adictedtohumping, Towelie, Cyrus Fett, Jean-françois SEMA, Kurt, max blo, Xenon Xue, JackJohnnyJim, Edward Ten Eyck, Chase Kwon, Inyoshu, Goober719, Chad Barnes, James Ming, vanditking, kripitonga, Rizzi, nimin, OMAR LUCIANO, hannibal, Jo+Example, BrentBertram, eumelzocker, dxjaymz, L C, Dude\n\n\u003C\u002Fdetails>\n\n\u003C!-- SUPPORTERS-END -->\n\n## Release Notes\n\n### v1.0.3\n\n* **Custom Recipe Storage Path** - Added support for configuring a custom storage path for recipes, with migration support to move existing recipe data when changing locations.\n* **Wildcard Support for LM Text\u002FPrompt Nodes** - The LM `Text` node and `Prompt` node now support the new `\u002Fwildcard` command, with runtime wildcard expansion and support for dynamic prompt syntax for more flexible prompt construction.\n* **System Diagnostics (\"Doctor\")** - Added a new diagnostics feature to help surface environment and setup issues more clearly.\n* **User-State Backup Support** - Added backup support for user state, with accompanying UI and clearer backup scope messaging in Settings.\n* **Downloaded Status Visibility** - Added clearer downloaded-status UX so previously downloaded model versions are easier to recognize.\n* **Autocomplete Performance Improvements** - Fixed autocomplete performance issues to reduce tag-search overhead and improve responsiveness.\n\n### v1.0.2\n\n* **Model Download History Tracking** - LoRA Manager now keeps a history of downloaded model versions, allowing it to recognize whether a version has been downloaded before, even if it is no longer currently present in your library.\n* **Skip Previously Downloaded Model Versions** - Added a new setting, `Skip previously downloaded model versions`, to help avoid downloading model versions you have already downloaded in the past.\n* **LoRA Stack Combiner Trigger Words Fix** - Fixed an issue where trigger word updates from `LORA_STACK` inputs were not propagated correctly through the LoRA Stack Combiner node.\n* **CivitAI Example Image Compatibility** - Improved support for CivitAI CDN subdomains so example images load more reliably.\n\n### v1.0.1\n\n* **Batch Recipe Import** - Import recipes from multiple URLs or directories simultaneously with optimized concurrency.\n* **Bulk Download Missing LoRAs** - New bulk action for recipes: select multiple recipes and download all missing LoRAs for the selected recipes in one operation.\n* **Import-Only Recipe Option** - Save recipe metadata without downloading missing LoRAs, allowing you to save interesting recipes for later and download dependencies when needed.\n* **Editable Recipe Prompts** - Edit recipe prompts directly in the recipe detail modal.\n* **Checkpoint Loader LM Node** - Behaves like ComfyUI's built-in Load Checkpoint node, with the added ability to load checkpoints from Extra Folder Paths.\n* **UNET Loader LM Node** - Behaves like ComfyUI's built-in Load Diffusion Model node, with support for loading from Extra Folder Paths and GGUF format (requires ComfyUI-GGUF custom node).\n* **LoRA Stack Combiner Node** - Merge two LoRA stacks into one. For example: use separate Randomizers for character and style LoRAs, then combine before applying.\n* **LoRA Pool Regex Filtering** - Filter which LoRAs enter the pool using custom regex patterns for include\u002Fexclude rules.\n* **Dynamic Base Model Types** - Base model types are now fetched dynamically from Civitai API, keeping them synchronized with the latest available models.\n* **Prompt Autocomplete Enhancements** - Tab key acceptance, configurable behavior, and improved multi-word tag matching.\n* **Download Base Model Exclusions** - Exclude specific base models from download operations when you only want certain model types.\n* **Mature Blur Threshold Setting** - Configure blur levels (`PG13` \u002F `R` \u002F `X` \u002F `XXX`, default `R+`) for mature content previews.\n* **Experimental: Nunchaku Qwen LoRA Support** - Experimental support for loading and applying LoRAs to Nunchaku quantized Qwen-Image models.\n* **Bug Fixes & UX Improvements** - Various fixes for a smoother workflow.\n\n### v1.0.0\n* **Extra Folder Paths Support** - Added support for additional model root paths exclusive to LoRA Manager. This allows loading LoRAs from extra locations outside ComfyUI's standard folders, helping avoid performance issues when working with large model libraries.\n* **Settings UI Overhaul** - Redesigned the Settings interface with a more organized layout, making it easier to find and configure application settings.\n* **Lazy Hash Computation** - Implemented lazy hash calculation for large model files (checkpoints and diffusion models). Hashes are now computed only when strictly necessary, minimizing redundant disk I\u002FO and significantly accelerating application initialization.\n* **Milestone & Supporter Recognition** - Updated the Supporter window to show appreciation for all project supporters as this v1.0.0 milestone is reached. Great thanks to the community for the ongoing support!\n* **Bug Fixes & UX Enhancements** - Various bug fixes and user experience improvements for a smoother workflow.\n\n### v0.9.16\n* **Duplicate Detection Enhancement** - The model duplicates mode now respects filter configurations, making it easier to find duplicate groups within specific filtered results.\n* **Tag Logic Toggle** - Added OR\u002FAND toggle for include tags filtering in the filters panel, providing more flexible tag-based model searches.\n* **Metadata Refresh Skip Paths** - New setting to exclude specific paths from metadata refresh operations. Models under these paths will be skipped when fetching metadata from remote sources.\n* **Dynamic Trigger Words in Prompt Node** - Prompt node now supports dynamic numbers of trigger word inputs for greater flexibility.\n* **Early Access Updates** - Model updates now display Early Access information, with a new setting to ignore Early Access updates if desired.\n* **LM Civitai Extension Integration** - Added integration with the LM Civitai Extension. Clicking the download button in model updates now sends downloads to the extension's download queue for seamless one-click downloads.\n\n### v0.9.15\n* **Filter Presets** - Save filter combinations as presets for quick switching and reapplication.\n* **Bug Fixes** - Fixed various bugs for improved stability.\n\n### v0.9.14\n* **LoRA Cycler Node** - Introduced a new LoRA Cycler node that enables iteration through specified LoRAs with support for repeat count and pause iteration functionality. Refer to the new \"Lora Cycler\" template workflow for concrete example.\n* **Enhanced Prompt Node with Tag Autocomplete** - Enhanced the Prompt node with comprehensive tag autocomplete based on merged Danbooru + e621 tags. Supports tag search and autocomplete functionality. Implemented a command system with shortcuts like `\u002Fcharacter` or `\u002Fartist` for category-specific tag searching. Added `\u002Fac` or `\u002Fnoac` commands to quickly enable or disable autocomplete. Refer to the \"Lora Manager Basic\" template workflow in ComfyUI -> Templates -> ComfyUI-Lora-Manager for detailed tips.\n* **Bug Fixes & Stability** - Addressed multiple bugs and improved overall stability.\n\n### v0.9.12\n* **LoRA Randomizer System** - Introduced a comprehensive LoRA randomization system featuring LoRA Pool and LoRA Randomizer nodes for flexible and dynamic generation workflows.\n* **LoRA Randomizer Template** - Refer to the new \"LoRA Randomizer\" template workflow for detailed examples of flexible randomization modes, lock & reuse options, and other features.\n* **Recipe Folders** - Introduced a folder system for the Recipes page, allowing users to freely organize recipes just like they do with models.\n* **Recipe Bulk Operations** - Added bulk mode support for batch moving, deleting, and setting base models for selected recipes with intuitive controls like click-and-drag selection, drag-to-folder, and Ctrl+A (Select All).\n* **Prompt Search & Sorting** - Search recipes by prompt content and sort by Recipe Name, Imported Date, or LoRA Count for better browsing.\n* **Recipe Favorites** - Mark specific recipes as favorites for quick access.\n* **Video Recipe Support** - Enabled support for video recipes (import via LM extension or URL; video file import not supported).\n* **Performance Improvements** - Fixed performance issues for dramatically improved startup and loading speed. After first scan, subsequent loads are instant regardless of collection size.\n* **ComfyUI Nodes 2.0 Support** - Basic support for ComfyUI Nodes 2.0.\n\n### v0.9.10\n* **Smarter Update Matching** - Users can now choose to check and group updates by matching base model only or with no base-model constraint; version lists also support toggling between same-base versions or all versions.\n* **Flexible Tag Filtering** - The filter panel now supports tag exclusion: click a tag to include, click again to exclude, and click a third time to clear, enabling stronger and more flexible tag filters.\n* **License Visibility & Controls** - Model detail headers and ComfyUI preview popups now show Civitai license icons. The filter panel gains license include\u002Fexclude options, and a new global context menu action, \"Refresh license metadata,\" fetches missing license data.\n* **Recipe Improvements** - Recipes now allow importing with zero LoRAs, and recipe detail pages show the related checkpoint for easier reference.\n* **Better ZIP Downloads** - When downloading models packaged in ZIPs, model files are extracted into the target model folder; ZIPs containing multiple model files (e.g., WanVideo high\u002Flow LoRA pairs) are added as separate models.\n* **Template Workflow Update** - Refreshed the \"Illustrious Pony Example\" template workflow with usage guidance for each LoRA Manager node.\n* **Bug Fixes & Stability** - General fixes and stability improvements.\n\n### v0.9.9\n* **Check for Updates Feature** - Users can now check for updates for all models or selected models in bulk mode. Models with available updates will display an \"update available\" badge on their model card, and users can filter to show only models with updates.\n* **Model Versions Management** - Added a new Versions tab in the model modal that centralizes all versions of a model, providing download, delete, and ignore update functions.\n* **Send Checkpoint to ComfyUI** - Users can now click the send button on a checkpoint card to send the checkpoint directly to the current workflow's checkpoint or diffusion model loader node in ComfyUI.\n* **Customizable Model Card Display** - Added a new setting that allows users to choose whether to display the model name or filename on model cards.\n* **New Path Template Placeholders** - Added new path template placeholders: `{model_name}` and `{version_name}` for more flexible organization.\n* **ComfyUI Auto Path Correction Setting** - Added a new setting within ComfyUI to enable or disable the auto path correction feature.\n\n### v0.9.8\n* **Full CivArchive API Support** - Added complete support for the CivArchive API as a fallback metadata source beyond Civitai API. Models deleted from Civitai can now still retrieve metadata through the CivArchive API.\n* **Download Models from CivArchive** - Added support for downloading models directly from CivArchive, similar to downloading from Civitai. Simply click the Download button and paste the model URL to download the corresponding model.\n* **Custom Priority Tags** - Introduced Custom Priority Tags feature, allowing users to define custom priority tags. These tags will appear as suggestions when editing tags or during auto organization\u002Fdownload using default paths, providing more precise and controlled folder organization. [Guide](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Fwiki\u002FPriority-Tags-Configuration-Guide)\n* **Drag and Drop Tag Reordering** - Added drag and drop functionality to reorder tags in the tags edit mode for improved usability.\n* **Download Control in Example Images Panel** - Added stop control in the Download Example Images Panel for better download management.\n* **Prompt (LoraManager) Node with Autocomplete** - Added new Prompt (LoraManager) node with autocomplete feature for adding embeddings.\n* **Lora Manager Nodes in Subgraphs** - Lora Manager nodes now support being placed within subgraphs for more flexible workflow organization.\n\n### v0.9.6\n* **Metadata Archive Database Support** - Added the ability to download and utilize a metadata archive database, enabling access to metadata for models that have been deleted from CivitAI.\n* **App-Level Proxy Settings** - Introduced support for configuring a global proxy within the application, making it easier to use the manager behind network restrictions.\n* **Bug Fixes** - Various bug fixes for improved stability and reliability.\n\n### v0.9.2\n* **Bulk Auto-Organization Action** - Added a new bulk auto-organization feature. You can now select multiple models and automatically organize them according to your current path template settings for streamlined management.\n* **Bug Fixes** - Addressed several bugs to improve stability and reliability.\n\n### v0.9.1\n* **Enhanced Bulk Operations** - Improved bulk operations with Marquee Selection and a bulk operation context menu, providing a more intuitive, desktop-application-like user experience.\n* **New Bulk Actions** - Added bulk operations for adding tags and setting base models to multiple models simultaneously.\n\n### v0.9.0\n* **UI Overhaul for Enhanced Navigation** - Replaced the top flat folder tags with a new folder sidebar and breadcrumb navigation system for a more intuitive folder browsing and selection experience.\n* **Dual-Mode Folder Sidebar** - The new folder sidebar offers two display modes: 'List Mode,' which mirrors the classic folder view, and 'Tree Mode,' which presents a hierarchical folder structure for effortless navigation through nested directories.\n* **Internationalization Support** - Introduced multi-language support, now available in English, Simplified Chinese, Traditional Chinese, Spanish, Japanese, Korean, French, Russian, and German. Feedback from native speakers is welcome to improve the translations.\n* **Automatic Filename Conflict Resolution** - Implemented automatic file renaming (`original name + short hash`) to prevent conflicts when downloading or moving models.\n* **Performance Optimizations & Bug Fixes** - Various performance improvements and bug fixes for a more stable and responsive experience.\n\n[View Update History](.\u002Fupdate_logs.md)\n\n---\n\n## **⚠ Important Note**: To use the CivitAI download feature, you'll need to:\n\n1. Get your CivitAI API key from your profile settings\n2. Add it to the LoRA Manager settings page\n3. Save the settings\n\n---\n\n## Key Features\n\n- 🚀 **High Performance**\n  - Fast model loading and browsing\n  - Smooth scrolling through large collections\n  \n- 🌐 **Rich Model Integration**\n  - Direct download from CivitAI\n  - Preview images and videos\n  - Model descriptions and version selection\n  - Trigger words at a glance\n  - One-click workflow integration with preset values\n  \n- 🔄 **Checkpoint Management**\n  - Scan and organize checkpoint models\n  - Filter and search your collection\n  - View and edit metadata\n  - Clean up and manage disk space\n  \n- 🧩 **LoRA Recipes**\n  - Save and share favorite LoRA combinations\n  - Preserve generation parameters for future reference\n  - Quick application to workflows\n  - Import\u002Fexport functionality for community sharing\n  \n- 💻 **User Friendly**\n  - One-click access from ComfyUI menu\n  - Context menu for quick actions\n  - Custom notes and usage tips\n  - Multi-folder support\n  - Configurable mature blur threshold (`PG13` \u002F `R` \u002F `X` \u002F `XXX`, default `R+`)\n    - Example: setting threshold to `PG13` blurs `PG13`, `R`, `X`, and `XXX` previews when blur is enabled\n  - Visual progress indicators during initialization\n\n---\n\n## Installation\n\n### Option 1: **ComfyUI Manager** (Recommended for ComfyUI users)\n\n1. Open **ComfyUI**.\n2. Go to **Manager > Custom Node Manager**.\n3. Search for `lora-manager`.\n4. Click **Install**.\n\n### Option 2: **Portable Standalone Edition** (No ComfyUI required)\n\n1. Download the [Portable Package](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Freleases\u002Fdownload\u002Fv1.0.0\u002Flora_manager_portable.7z)\n2. Copy the provided `settings.json.example` file to create a new file named `settings.json` in `comfyui-lora-manager` folder.\n3. Edit the new `settings.json` to include your correct model folder paths and CivitAI API key\n   - Set `\"use_portable_settings\": true` if you want the configuration to remain inside the repository folder instead of your user settings directory.\n4. Run run.bat\n    - To change the startup port, edit `run.bat` and modify the parameter (e.g. `--port 9001`)\n\n### Option 3: **Manual Installation**\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager.git\ncd ComfyUI-Lora-Manager\npip install -r requirements.txt\n```\n\n## Usage\n\n1. There are two ways to access the LoRA manager:\n   - Click the \"Launch LoRA Manager\" button in the ComfyUI menu\n   - Visit http:\u002F\u002Flocalhost:8188\u002Floras directly\n2. From the interface, you can:\n   - Browse and organize your LoRA models\n   - Download models directly from CivitAI\n   - Automatically fetch or manually set preview images\n   - View and copy trigger words associated with each LoRA\n   - Add personal notes and usage tips\n3. To use LoRAs in your workflow:\n   - Add the \"Lora Loader (LoraManager)\" node to your workflow\n   - Select a LoRA in the manager interface\n   - Click copy button or use right-click menu \"Copy LoRA syntax\"\n   - Paste into the Lora Loader node's text input\n   - The node will automatically apply preset strength and trigger words\n\n### Wildcards for TextLM \u002F PromptLM\n\n`Text (LoraManager)` and `Prompt (LoraManager)` support `\u002Fwildcard` autocomplete plus runtime wildcard expansion.\n\n- Wildcard files live in `{settings folder}\u002Fwildcards\u002F`\n- When you type `\u002Fwildcard` and no wildcard files exist yet, the autocomplete dropdown shows the exact folder path and lets you open it\n- Supported formats: `.txt`, `.yaml`, `.yml`, `.json`\n\nFormat rules:\n\n- `wildcards\u002Fanimals\u002Fcat.txt` becomes `__animals\u002Fcat__`\n- `.txt` files use one option per line\n- YAML \u002F JSON files use nested keys that end in string arrays\n\nExamples:\n\n```txt\n# wildcards\u002Fcolor.txt\nred\nblue\ngreen\n```\n\nUse it as `__color__`.\n\n```yaml\n# wildcards\u002Fcolors.yaml\npalette:\n  warm:\n    - red\n    - orange\n```\n\nUse it as `__palette\u002Fwarm__`.\n\n### Filename Format Patterns for Save Image Node\n\nThe Save Image Node supports dynamic filename generation using pattern codes. You can customize how your images are named using the following format patterns:\n\n#### Available Pattern Codes\n\n- `%seed%` - Inserts the generation seed number\n- `%width%` - Inserts the image width\n- `%height%` - Inserts the image height\n- `%pprompt:N%` - Inserts the positive prompt (limited to N characters)\n- `%nprompt:N%` - Inserts the negative prompt (limited to N characters)\n- `%model:N%` - Inserts the model\u002Fcheckpoint name (limited to N characters)\n- `%date%` - Inserts current date\u002Ftime as \"yyyyMMddhhmmss\"\n- `%date:FORMAT%` - Inserts date using custom format with:\n  - `yyyy` - 4-digit year\n  - `yy` - 2-digit year\n  - `MM` - 2-digit month\n  - `dd` - 2-digit day\n  - `hh` - 2-digit hour\n  - `mm` - 2-digit minute\n  - `ss` - 2-digit second\n\n#### Examples\n\n- `image_%seed%` → `image_1234567890`\n- `gen_%width%x%height%` → `gen_512x768`\n- `%model:10%_%seed%` → `dreamshape_1234567890`\n- `%date:yyyy-MM-dd%` → `2025-04-28`\n- `%pprompt:20%_%seed%` → `beautiful landscape_1234567890`\n- `%model%_%date:yyMMdd%_%seed%` → `dreamshaper_v8_250428_1234567890`\n\nYou can combine multiple patterns to create detailed, organized filenames for your generated images.\n\n### Standalone Mode\n\nYou can now run LoRA Manager independently from ComfyUI:\n\n1. **For ComfyUI users**:\n   - Launch ComfyUI with LoRA Manager at least once to initialize the necessary path information in the `settings.json` file located in your user settings folder (see paths above).\n   - Make sure dependencies are installed: `pip install -r requirements.txt`\n   - From your ComfyUI root directory, run:\n     ```bash\n     python custom_nodes\\comfyui-lora-manager\\standalone.py\n     ```\n   - Access the interface at: `http:\u002F\u002Flocalhost:8188\u002Floras`\n   - You can specify a different host or port with arguments:\n     ```bash\n     python custom_nodes\\comfyui-lora-manager\\standalone.py --host 127.0.0.1 --port 9000\n     ```\n\n2. **For non-ComfyUI users**:\n   - Copy the provided `settings.json.example` file to create a new file named `settings.json`. Update the API key, optional language, and folder paths only—the library registry is created automatically when LoRA Manager starts.\n   - Edit `settings.json` to include your correct model folder paths and CivitAI API key (you can leave the defaults until ready to configure them)\n   - Enable portable mode by setting `\"use_portable_settings\": true` if you prefer LoRA Manager to read and write the `settings.json` located in the project directory.\n   - Install required dependencies: `pip install -r requirements.txt`\n   - Run standalone mode:\n     ```bash\n     python standalone.py\n     ```\n   - Access the interface through your browser at: `http:\u002F\u002Flocalhost:8188\u002Floras`\n\n   > **Note:** Existing installations automatically migrate the legacy `settings.json` from the plugin folder to the user settings directory the first time you launch this version.\n\nThis standalone mode provides a lightweight option for managing your model and recipe collection without needing to run the full ComfyUI environment, making it useful even for users who primarily use other stable diffusion interfaces.\n\n## Testing & Coverage\n\n### Backend\n\nInstall the development dependencies and run pytest with coverage reports:\n\n```bash\npip install -r requirements-dev.txt\nCOVERAGE_FILE=coverage\u002Fbackend\u002F.coverage pytest \\\n  --cov=py \\\n  --cov=standalone \\\n  --cov-report=term-missing \\\n  --cov-report=html:coverage\u002Fbackend\u002Fhtml \\\n  --cov-report=xml:coverage\u002Fbackend\u002Fcoverage.xml \\\n  --cov-report=json:coverage\u002Fbackend\u002Fcoverage.json\n```\n\nHTML, XML, and JSON artifacts are stored under `coverage\u002Fbackend\u002F` so you can inspect hot spots locally or from CI artifacts.\n\n### Frontend\n\nRun the Vitest coverage suite to analyze widget hot spots:\n\n```bash\nnpm run test:coverage\n```\n\n---\n\n## Documentation\n\n- **[metadata.json Schema Documentation](docs\u002Fmetadata-json-schema.md)** — Complete reference for the `.metadata.json` sidecar file format, including all fields, types, and examples for LoRA, Checkpoint, and Embedding models.\n\n---\n\n## Contributing\n\nThank you for your interest in contributing to ComfyUI LoRA Manager! As this project is currently in its early stages and undergoing rapid development and refactoring, we are temporarily not accepting pull requests.\n\nHowever, your feedback and ideas are extremely valuable to us:\n- Please feel free to open issues for any bugs you encounter\n- Submit feature requests through GitHub issues\n- Share your suggestions for improvements\n\nWe appreciate your understanding and look forward to potentially accepting code contributions once the project architecture stabilizes.\n\n---\n\n## Credits\n\nThis project has been inspired by and benefited from other excellent ComfyUI extensions:\n\n- [ComfyUI-QwenImageLoraLoader](https:\u002F\u002Fgithub.com\u002Fussoewwin\u002FComfyUI-QwenImageLoraLoader) - For the experimental Nunchaku Qwen-Image LoRA support\n- [ComfyUI-SaveImageWithMetaData](https:\u002F\u002Fgithub.com\u002Fnkchocoai\u002FComfyUI-SaveImageWithMetaData) - For the image metadata functionality\n- [rgthree-comfy](https:\u002F\u002Fgithub.com\u002Frgthree\u002Frgthree-comfy) - For the lora loader functionality\n\n---\n\n## ☕ Support\n\nIf you find this project helpful, consider supporting its development:\n\n[![ko-fi](https:\u002F\u002Fko-fi.com\u002Fimg\u002Fgithubbutton_sm.svg)](https:\u002F\u002Fko-fi.com\u002Fpixelpawsai)\n\n[![Patreon](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBecome%20a%20Patron-F96854.svg?style=for-the-badge&logo=patreon&logoColor=white)](https:\u002F\u002Fpatreon.com\u002FPixelPawsAI)\n\nWeChat: [Click to view QR code](https:\u002F\u002Fraw.githubusercontent.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Fmain\u002Fstatic\u002Fimages\u002Fwechat-qr.webp)\n\n## 💬 Community\n\nJoin our Discord community for support, discussions, and updates:\n[Discord Server](https:\u002F\u002Fdiscord.gg\u002FvcqNrWVFvM)\n\n---\n## Star History\n\n[![Star History Chart](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_cb3548a2d67b.png)](https:\u002F\u002Fstar-history.com\u002F#willmiao\u002FComfyUI-Lora-Manager&Date)\n","# ComfyUI LoRA 管理器\n\n> **用 ComfyUI 的终极 LoRA 伴侣彻底革新你的工作流程！**\n\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fdiscord\u002F1346296675538571315?color=7289DA&label=Discord&logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002FvcqNrWVFvM)\n[![发布版本](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Fwillmiao\u002FComfyUI-Lora-Manager?include_prereleases&color=blue&logo=github)](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Freleases)\n[![发布日期](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Frelease-date\u002Fwillmiao\u002FComfyUI-Lora-Manager?color=green&logo=github)](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Freleases)\n\n这是一套全面的工具集，可简化在 ComfyUI 中组织、下载和应用 LoRA 模型的过程。凭借配方管理、检查点整理以及一键式工作流集成等强大功能，使用模型将变得更加快速、流畅且轻松。可通过以下地址访问界面：`http:\u002F\u002Flocalhost:8188\u002Floras`\n\n![界面预览](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_dfbb1891894f.png)\n\n## 📺 教程：一键式 LoRA 集成\n观看本快速教程，了解如何使用全新的一键式 LoRA 集成功能：\n\n[![一键式 LoRA 集成教程](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_07a60e6715d8.jpg)](https:\u002F\u002Fyoutu.be\u002FhvKw31YpE-U)\n\n## 🌐 浏览器扩展\n借助我们的配套浏览器扩展，提升你在 Civitai 上的浏览体验吧！查看你已拥有的模型、一键下载新模型，并高效管理你的下载内容。\n\n![LM Civitai 扩展预览](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_39e4270dea80.png)\n\n\u003Cdiv>\n  \u003Ca href=\"https:\u002F\u002Fchromewebstore.google.com\u002Fdetail\u002Flm-civitai-extension\u002Fcapigligggeijgmocnaflanlbghnamgm?utm_source=item-share-cb\" style=\"display: inline-block; background-color: #4285F4; color: white; padding: 8px 16px; text-decoration: none; border-radius: 4px; font-weight: bold; margin: 10px 0;\">\n    \u003Cimg src=\"https:\u002F\u002Fwww.google.com\u002Fchrome\u002Fstatic\u002Fimages\u002Fchrome-logo.svg\" width=\"20\" style=\"vertical-align: middle; margin-right: 8px;\"> 从 Chrome 网上应用店获取扩展\n  \u003C\u002Fa>\n\u003C\u002Fdiv>\n\n\u003Cdiv id=\"firefox-install\" class=\"install-ok\">\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fwillmiao\u002Flm-civitai-extension-firefox\u002Freleases\u002Flatest\u002Fdownload\u002Fextension.xpi\">📦 安装 Firefox 扩展（经 Mozilla 审核并认证）\u003C\u002Fa>\u003C\u002Fdiv>\n\n📚 [了解更多：完整教程](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Fwiki\u002FLoRA-Manager-Civitai-Extension-(Chrome-Extension))\n\n---\n\n## 🤝 支持者表彰\n\n我由衷感谢每一位支持本项目开发的朋友。看到这么多人认可我的工作，是我所能获得的最佳动力。你们的信任与支持对打造如今这款工具起到了至关重要的作用，借此机会我要特别向那些帮助实现这一切的人表示感谢。\n\n\u003C!-- SUPPORTERS-START -->\n\n### 🌟 特别致谢\n\n**dispenser**, **EbonEagle**, **DanielMagPizza**, **Scott R**\n\n### 💖 支持者（666 位）\n\n\u003Cdetails>\n\u003Csummary>点击查看所有了不起的支持者\u003C\u002Fsummary>\n\u003Cbr>\n\n失眠艺术设计、megakirbs、布伦诺克、2018cfh、W+K+White、wackop、塔坎、卡尔G、$MetaSamsara、itismyelement、onesecondinosaur、stone9k、罗森塔尔、弗朗西斯科·塔蒂斯、安德鲁·威尔逊、格雷布什、Gooohokrbe、瑞奇·卡特、韩宗元、OldBones、VantAI、runte3221、FreelancerZ、埃德加·特赫达、利亚姆·麦克道格尔、弗雷泽·克罗斯、多态不定型、伯迪、马克·威芬、豪尔赫·胡斯尼、基巴、Skalabananen、雷诺·林、sig、克里斯蒂安·伯恩、DM、Sen314、艾斯托根、J\\B\u002F 8r0wns0n、Snaggwort、阿尔莱基诺·希昂、查尔斯·布莱克莫尔、罗布·威廉姆斯、ClockDaemon、KD、Omnidex、泰勒·特鲁布肖恩、Release Cabrakan、Tobi_Swagg、SG、carozzz、詹姆斯·杜利、zenbound、秃鹫、jmack、马克·科内格里奥、讽刺标签、Cosmosis、iamresist、RedrockVP、沃尔芬、FloPro4Sho、詹姆斯·托德、史蒂文·普费弗、蒂姆、Lisster、迈克尔·王、Illrigger、汤姆·科里根、杰基·王、fnkylove、朱利安·V、史蒂文·欧文斯、Yushio、Vik71it、Echo、Lilleman、罗伯特·斯泰西、PM、托德·凯克、Mozzel、银杏叶、Sterilized、BadassArabianMofo、帕斯卡·达勒、quarz、格雷格、Penfore、JSST、esthe、lmsupporter、IamAyam、wfpearl、Baekdoosixt、乔纳森·罗斯、杰克·B·敏捷、Nazono_hito、梅尔维尔·帕里什、丹尼尔·鸽子、Lustre、JW Sin、contrite831、亚历克斯、bh、被没收的Zyra、马龙·丹尼尔斯、Starkselle、亚伦·布劳尔、LacesOut!、greebles、亚当·肖、Tee Gee、安东尼·里佐、塔雷克·赫尔米、M Postkasse、ASLPro3D、雅各布·霍勒尔、FinalyFree、Weasyl、蒂米、约翰尼、科里·帕扎、塔克、冈萨洛·安德烈·阿连德斯·洛佩斯、扎克·贡瑟、大红、whudunit、卢克·乔布、dl0901dm、菲利普·亨佩尔、corde、尼克·沃克、lh qwe、Bishoujoker、conner、aai、不列颠人海尔布鲁恩、托里、wildnut、亮眼睛公主、AbstractAss、费利佩·多斯桑托斯、ViperC、让·贾伦、亚历山大·武伊奇克、AM Kuro、马库斯、S Sang、卡尔P、Akira_HentAI、MagnaInsomnia、戈登·科尔、yuxz69、道格拉斯·加斯帕尔、AlexDuKaNa、乔治、andrew.tappan、dw、N\u002FA、The Spawn、菲尔、graysock、Greenmoustache、zounic、fancypants、Digital、JaxMax、takyamtom、奚明 刘、Jwk0205、Bro Xie、준희 김、batblue、carey6409、Olive、太郎 ゲーム、某个叫巴里的家伙、马克斯·马克伦德、友博弘·巴巴、大卫·奥尔特加、AELOX、Nicfit23、Noora、wamekukyouzin、drum matthieu、Dogmaster、马特·温泽尔、Mattssn、Lex Song、约翰·保存为、克里斯托弗·米歇尔、塞尔日·贝肯坎普、吉米·莱德贝特、LeoZero、安东尼·蓬特斯、ApathyJones、nahinahi9、达斯汀·陈、dan、Yaboi、Mouthlessman、Steam Steam、达蒙·坎利夫、CryptoTraderJK、Davaitamin、御宅族fra、Ran C、tedcor、Fotek Design、亚当·泰勒、Weird_With_A_Beard、MadSpin、Pozadine1、Qarob、AIGooner、inbijiburu、Luc、ProtonPrince、DiffDuck、elu3199、尼克“Loadstone”D、Hasturkun、乔恩·沙曼、Ubivis、CloudValley、thesoftwaredruid、wundershark、mr_dinosaur、linnfrey、Gamalonia、Vir、Pkrsky、Joboshy、波西米亚下士、Dan、约瑟夫·兰兹尔、塞思·克里斯滕森、格里芬·达尔伯格、德拉文T、yer fey、Error_Rule34_Not_found、杰拉尔德·韦利、Roslynd、Geolog、jinxedx、Neco28、Aquatic Coffee、Dankin、ethanfel、克里斯蒂安·巴斯克斯、弗兰克·尼蒂、Magic Noob、Focuschannel、道格·彼得森、杰夫、布鲁斯、凯文·约翰·鸭、安东尼·法克斯兰德、凯文·克里斯托弗、Ouro Boros、Blackfish95、dd、保罗·克罗尔、MiraiKuriyamaSy、分号排水管、Thesharingbrother、Bas Imagineer、Pat Hen、约翰·斯塔森、ResidentDeviant、Nihongasuki、JC、Prompt Pirate、uwutismxd、诱饵、Tyrswood、Ray Wing、Ranzitho、Gus、地狱的禄、MJG、大卫·拉瓦莱、ae、Tr4shP4nda、WRL_SPR、capn、约瑟夫、米尔科·卡祖拉、dan、Piccio08、kumakichi、cppbel、starbugx、Moon Knight、몽타주、Kland、zenobeus、Jackthemind、ryoma、Stryker、raf8osz、ElitaSSJ4、blikkies、克里斯、布赖恩M、Nerezza、sanborondon、泰勒·放克、aezin、Thought2Form、jcay015、凯文·皮科、埃里克·洛佩斯、Shock Shockor、马特奥·库里奇、Goldwaters、Zude、Eris3D、m、皮尔斯·麦克布莱德、乔舒亚·格雷、凯勒、米科·海米拉、aRtFuL_DodGeR、杰米·奥格莱特里、a _、詹姆斯·科尔曼、CrimsonDX、Martial、battu、埃米尔·安德森、查德Idk、DarkSunset、比利·格拉德基、优治·金子、Probis、杜尚·里班、ItsGeneralButtNaked、乔丹·肖、Rops Alot、萨姆、sjon kreutz、Nimess、SRDB、Ace Ventura、g unit、Youguang、Metryman55、andrewzpong、FrxzenSnxw、BossGame、lrdchs、momokai、Hailshem、kudari、娜奥米·黑尔·丹奇、dc7431、ken、Inversity、AIVORY3D、epicgamer0020690、乔舒亚·波拉塔、keemun、SuBu、RedPIXel、Kevinj、Wind、Nexus、Ramneek“Guy”Ashok、squid_actually、Nat_20、爱德华·周、kyoumei、RadStorm04、JohnDoe42054、比利希尔、emyth、chriphost、KitKatM、socrasteeze、ResidentDeviant、gzmzmvp、Welkor、约翰·马丁、理查德、安德鲁、罗伯特·韦格蒙德、Littlehuggy、moranqianlong、格雷戈里·科热米亚克、mrjuan、布赖恩·布耶、Sadlip、春·约图、埃里克·惠特尼、乔伊·卡拉汉、伊万·塔迪奇、迈克·西蒙、Morgandel、凯隆·马汉、Matura Arbeit、诺亚、雅各布·麦克丹尼尔、X、斯隆·斯蒂迪、TBitz33、匿名dkjglfleeoeldldldlkf、Temikus、Artokun、迈克尔·泰勒、SendingRavens、德里克·贝克、迈克·安东尼·斯科特、阿提拉·贝尔克·佩克杜亚尔、迈克·多切蒂、内森、Decx _、保罗·哈茨尤克、elitassj、雅各布·温特、Distortik、大卫、Meilo、Pen Bouryoung、四糸凜音、shinonomeiro、Snille、马尔滕·阿尔伯斯、khanh duy、xybrightsummer、jreedatchison、PhilW、Tree Tagger、Janik、克罗克特、Cruel、MRBlack、米切尔·罗布森、Kiyoe、humptynutz、michael.isaza、Kalnei、Whitepinetrader、OrganicArtifact、斯科特、MudkipMedkitz、deanbrian、POPPIN、亚历克斯·沃特曼、科迪、Raku、smart.edge5178、emadsultan、InformedViewz、CHKeeho80、Bubbafett、leaf、Menard、Skyfire83、亚当·莱因哈特、D、Pitpe11、TheD1rtyD03、moonpetal、SomeDude、g9p0o、nanana、TheHolySheep、Monte Won、SpringBootisTrash、carsten、ikok、Buecyb99、4IXplr0r3r、dfklsjfkljslfjd、hayden、ahoystan、利兰·桑德斯、Wolfe7D1、Ink Temptation、鲍勃·巴克、edk、Kalli Core、Aeternyx、elleshar666、YOU SINWOO、ja s、道格·梅森、Kauffy、杰里米·汤森德、EpicElric、肖恩·沃茨、欧文·格沃兹德、约翰·J·莱恩汉、埃利奥特·E、托马斯·瓦纳、Theerat Jiramate、爱德华·肯尼迪、贾斯汀·布莱洛克、Devil Lude、尼克·卡格、凯文·斯托达德、杰克·多尔、Vane Holzer、psytrax、Ezokewn、hexxish、CptNeo、notedfakes、Maso、埃里克·凯彻姆、NICHOLAS BAXLEY、迈克·斯科特、凯文·华莱士、马修斯·库托、Saya、ChicRic、mercur、J C、Ed Wang、瑞安·普雷斯利·Ng、韦斯·西姆斯、Donor4115、伊夫·波泽瓦拉、Teriak47、就我一个人、拉夫·斯塔赫林、Вячеслав Маринин、Lyavph、菲利波·费拉里、可乐·马修、OniNoKen、伊恩·怀斯利、Zertens、NOHOW、Apo、nekotxt、choowkee、Clusters、ibrahim、Highlandrise、philcoraz、mztn、ImagineerNL、MrAcrtosSursus、al300680、pixl、Robin、chahknoir、马库斯·thronico、nd、keno94d、詹姆斯·梅尔策、巴特比、Renvertere、Rahuy、Hermann003、D、Foolish、RevyHiep、Captain_Swag、obkircher、gwyar、D、edgecase、Neoxena、mrmhalo、dg、马尔滕·哈姆斯、以色列、Muratoraccio、SelfishMedic、Ginnie、adderleighn、EnragedAntelope、艾伦+Cano、FeralOpticsAI、Pavlaki、generic404、马特乌什+科谢拉、道格+林托尔、Noor、Yorunai、Bula、quantenmecha、abattoirblues、杰森·纳什、比利男孩84、DarkRoast、zounik、letzte、Nasty+Hobbit、SgtFluffles、lrdchs2、Duk3+Rand0m、KUJYAKU、NathenChoi、托马斯+Reck、Larses、cocona、Coeur+de+cochon、大卫·申克、han b、Nico、香蕉乔、_ G3n、多诺万·詹金斯、JBsuede、迈克·艾德、beersandbacon、马克西米利安·派科、Invis、贾斯汀·休斯顿、Time Valentine、james、OrochiNights、迈克·朱、ACTUALLY_the_Real_Willem_Dafoe、gonzalo、Seraphy、Михал Михалыч、雨の心 落、马特、AllTimeNoobie、jumpd、约翰C、Rim、Dismem、Frogmilk、SPJ、Xan Dionysus、内森·李、Mewtora、Middo、Forbidden Atelier、布莱恩·鲁特科夫斯基、Adictedtohumping、Towelie、赛勒斯·费特、让-弗朗索瓦·SEMA、库尔特、max blo、Xenon Xue、杰克·约翰尼·吉姆、爱德华·腾·艾克、蔡·权、Inyoshu、Goober719、查德·巴恩斯、詹姆斯·明、vanditking、kripitonga、Rizzi、nimin、OMAR LUCIANO、汉尼拔、Jo+Example、布伦特·伯特拉姆、eumelzocker、dxjaymz、L C、Dude\n\n\u003C\u002Fdetails>\n\n\u003C!-- 支持者-结束 -->\n\n\n\n## 发行说明\n\n### v1.0.3\n\n* **自定义配方存储路径** - 新增支持为配方配置自定义存储路径，并在更改位置时提供迁移支持，以移动现有配方数据。\n* **LM 文本\u002F提示节点的通配符支持** - LM `文本`节点和`提示`节点现在支持新的`\u002Fwildcard`命令，具备运行时通配符扩展功能，并支持动态提示语法，从而实现更灵活的提示构建。\n* **系统诊断（“医生”）** - 新增诊断功能，帮助更清晰地发现环境和设置问题。\n* **用户状态备份支持** - 新增用户状态备份支持，配套UI以及在设置中更清晰的备份范围提示信息。\n* **已下载状态可见性** - 新增更清晰的已下载状态用户体验，使之前下载过的模型版本更容易识别。\n* **自动补全性能改进** - 修复了自动补全性能问题，以减少标签搜索开销并提高响应速度。\n\n### v1.0.2\n\n* **模型下载历史追踪** - LoRA管理器现在会保存已下载模型版本的历史记录，即使该版本当前不在您的库中，也能识别是否曾经下载过。\n* **跳过已下载过的模型版本** - 新增设置“跳过已下载过的模型版本”，帮助避免重复下载您过去已经下载过的模型版本。\n* **LoRA堆栈组合节点触发词修复** - 修复了一个问题：来自`LORA_STACK`输入的触发词更新无法正确传播到LoRA堆栈组合节点。\n* **CivitAI示例图片兼容性** - 改进了对CivitAI CDN子域名的支持，使示例图片加载更加可靠。\n\n### v1.0.1\n\n* **批量导入配方** - 可同时从多个URL或目录导入配方，并优化并发处理。\n* **批量下载缺失的LoRA** - 配方新增批量操作：选择多个配方后，可一次性下载所有选定配方中缺失的LoRA。\n* **仅导入配方选项** - 只保存配方元数据而不下载缺失的LoRA，方便您先保存感兴趣的配方，待需要时再下载依赖项。\n* **可编辑配方提示** - 直接在配方详情模态框中编辑配方提示。\n* **检查点加载LM节点** - 行为与ComfyUI内置的加载检查点节点相同，额外增加了从额外文件夹路径加载检查点的功能。\n* **UNET加载LM节点** - 行为与ComfyUI内置的加载扩散模型节点相同，支持从额外文件夹路径和GGUF格式加载（需使用ComfyUI-GGUF自定义节点）。\n* **LoRA堆栈组合节点** - 将两个LoRA堆栈合并为一个。例如：分别使用随机化器处理角色和风格的LoRA，然后再合并应用。\n* **LoRA池正则过滤** - 使用自定义的包含\u002F排除规则正则表达式来筛选进入池中的LoRA。\n* **基础模型类型动态获取** - 基础模型类型现在会从Civitai API动态获取，确保与最新可用模型保持同步。\n* **提示自动补全增强** - 支持Tab键确认、可配置行为，并改进了多词标签匹配功能。\n* **基础模型下载排除** - 当您只想下载特定类型的模型时，可以将某些基础模型排除在下载之外。\n* **成人内容模糊阈值设置** - 可配置成人内容预览的模糊级别（PG13 \u002F R \u002F X \u002F XXX，默认为R+）。\n* **实验性：Nunchaku Qwen LoRA支持** - 实验性支持加载并应用于Nunchaku量化的Qwen-Image模型的LoRA。\n* **Bug修复与UX改进** - 各种修复使工作流程更加顺畅。\n\n### v1.0.0\n* **额外文件夹路径支持** - 新增支持LoRA管理器专用的额外模型根路径。这允许从ComfyUI标准文件夹之外的其他位置加载LoRA，有助于避免在处理大型模型库时出现性能问题。\n* **设置界面全面改版** - 重新设计了设置界面，布局更加整洁，便于查找和配置应用设置。\n* **惰性哈希计算** - 对大型模型文件（检查点和扩散模型）实现了惰性哈希计算。哈希值仅在必要时才会计算，从而最大限度地减少冗余磁盘I\u002FO，并显著加快应用初始化速度。\n* **里程碑与支持者表彰** - 更新了支持者窗口，在达到v1.0.0里程碑之际，向所有项目支持者表示感谢。非常感谢社区一直以来的支持！\n* **Bug修复与UX提升** - 各种Bug修复和用户体验改进，使工作流程更加流畅。\n\n### v0.9.16\n* **重复检测增强** - 模型重复模式现在会尊重筛选器配置，使得在特定筛选结果中更容易找到重复组。\n* **标签逻辑切换** - 在筛选面板中新增OR\u002FAND切换开关，用于包含标签筛选，提供更灵活的基于标签的模型搜索功能。\n* **元数据刷新跳过路径** - 新增设置，可将特定路径排除在元数据刷新操作之外。当从远程源获取元数据时，这些路径下的模型将被跳过。\n* **提示节点中的动态触发词** - 提示节点现在支持动态数量的触发词输入，以获得更大的灵活性。\n* **抢先体验更新** - 模型更新现在会显示抢先体验信息，并新增设置以忽略抢先体验更新（如需）。\n* **LM Civitai扩展集成** - 新增与LM Civitai扩展的集成。点击模型更新中的下载按钮后，下载任务将发送至该扩展的下载队列，实现无缝的一键下载。\n\n### v0.9.15\n* **筛选器预设** - 可将筛选组合保存为预设，以便快速切换和重新应用。\n* **Bug修复** - 修复了各种Bug，提升了稳定性。\n\n### v0.9.14\n* **LoRA循环节点** - 引入了新的LoRA循环节点，可在指定的LoRA之间进行迭代循环，并支持重复次数和暂停迭代功能。具体示例请参考新的“Lora Cycler”模板工作流。\n* **增强型提示节点与标签自动补全** - 提升了提示节点的功能，基于Danbooru + e621标签的合并数据提供了全面的标签自动补全。支持标签搜索和自动补全功能。实现了命令系统，例如使用`\u002Fcharacter`或`\u002Fartist`等快捷指令进行特定类别的标签搜索。新增`\u002Fac`或`\u002Fnoac`命令，可快速启用或禁用自动补全功能。详细技巧请参阅ComfyUI -> 模板 -> ComfyUI-Lora-Manager中的“Lora Manager Basic”模板工作流。\n* **Bug修复与稳定性提升** - 解决了多个Bug，并提高了整体稳定性。\n\n### v0.9.12\n* **LoRA 随机化系统** - 引入了全面的 LoRA 随机化系统，包含 LoRA 池和 LoRA 随机化节点，用于灵活且动态的生成工作流。\n* **LoRA 随机化模板** - 请参考新的“LoRA 随机化”模板工作流，了解灵活随机化模式、锁定与复用选项等详细示例。\n* **配方文件夹** - 在配方页面引入了文件夹系统，用户可以像管理模型一样自由地组织配方。\n* **配方批量操作** - 新增批量模式支持，可通过直观的控件（如点击拖拽选择、拖拽到文件夹、Ctrl+A 全选）对选定的配方进行批量移动、删除和设置基础模型。\n* **提示词搜索与排序** - 可按提示词内容搜索配方，并按配方名称、导入日期或 LoRA 数量排序，以提升浏览体验。\n* **配方收藏** - 将特定配方标记为收藏，方便快速访问。\n* **视频配方支持** - 支持视频配方（可通过 LM 扩展或 URL 导入；不支持直接导入视频文件）。\n* **性能优化** - 修复了性能问题，显著提升了启动和加载速度。首次扫描后，无论收藏规模如何，后续加载都将即时完成。\n* **ComfyUI 节点 2.0 支持** - 基本支持 ComfyUI 节点 2.0。\n\n### v0.9.10\n* **更智能的更新匹配** - 用户现在可以选择仅按基础模型匹配或不限制基础模型来检查和分组更新；版本列表还支持在同基础版本与所有版本之间切换。\n* **灵活的标签筛选** - 筛选面板现支持标签排除功能：单击标签可包含，再次单击则排除，第三次单击则清空，从而实现更强、更灵活的标签筛选。\n* **许可证可见性与控制** - 模型详情页头部及 ComfyUI 预览弹窗现显示 Civitai 许可证图标。筛选面板新增许可证包含\u002F排除选项，同时推出全新全局上下文菜单操作“刷新许可证元数据”，用于获取缺失的许可证信息。\n* **配方改进** - 配方现允许导入时不含任何 LoRA，且配方详情页会显示相关检查点，便于参考。\n* **更好的 ZIP 下载** - 下载打包为 ZIP 的模型时，模型文件将被解压至目标模型文件夹；包含多个模型文件的 ZIP（例如 WanVideo 的高低 LoRA 对）将作为独立模型添加。\n* **模板工作流更新** - 更新了“卓越小马示例”模板工作流，并为每个 LoRA Manager 节点提供了使用指南。\n* **Bug 修复与稳定性提升** - 进行了常规修复并提升了稳定性。\n\n### v0.9.9\n* **检查更新功能** - 用户现在可在批量模式下检查所有模型或选定模型的更新。有可用更新的模型将在其模型卡片上显示“有更新”徽章，用户还可筛选仅显示有更新的模型。\n* **模型版本管理** - 在模型模态窗口中新增“版本”选项卡，集中管理模型的所有版本，提供下载、删除和忽略更新等功能。\n* **将检查点发送至 ComfyUI** - 用户现在可点击检查点卡片上的发送按钮，直接将检查点发送至 ComfyUI 当前工作流中的检查点或扩散模型加载节点。\n* **自定义模型卡片显示** - 新增设置，允许用户选择在模型卡片上显示模型名称还是文件名。\n* **新路径模板占位符** - 新增 `{model_name}` 和 `{version_name}` 路径模板占位符，以实现更灵活的文件组织。\n* **ComfyUI 自动路径修正设置** - 在 ComfyUI 中新增设置，用于启用或禁用自动路径修正功能。\n\n### v0.9.8\n* **CivArchive API 完全支持** - 新增对 CivArchive API 的完整支持，作为 Civitai API 之外的备用元数据来源。即使模型已从 Civitai 删除，仍可通过 CivArchive API 获取元数据。\n* **从 CivArchive 下载模型** - 新增直接从 CivArchive 下载模型的功能，类似于从 Civitai 下载。只需点击“下载”按钮并粘贴模型 URL 即可下载相应模型。\n* **自定义优先级标签** - 引入自定义优先级标签功能，允许用户定义自定义优先级标签。这些标签将在编辑标签时或使用默认路径进行自动整理\u002F下载时作为建议显示，从而实现更精准、更可控的文件夹组织。[指南](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Fwiki\u002FPriority-Tags-Configuration-Guide)\n* **标签拖放重排** - 在标签编辑模式中新增拖放功能，以便更便捷地重新排列标签。\n* **示例图片面板中的下载控制** - 在“下载示例图片”面板中新增停止控制，以更好地管理下载。\n* **带自动补全的 Prompt (LoraManager) 节点** - 新增带有自动补全功能的 Prompt (LoraManager) 节点，用于添加嵌入向量。\n* **子图中的 Lora Manager 节点** - Lora Manager 节点现支持放置于子图中，以实现更灵活的工作流组织。\n\n### v0.9.6\n* **元数据归档数据库支持** - 新增下载和使用元数据归档数据库的功能，使用户能够访问已从 CivitAI 删除的模型的元数据。\n* **应用级代理设置** - 引入了在应用程序内配置全局代理的支持，便于在网络限制环境下使用该管理器。\n* **Bug 修复** - 修复了多项 bug，以提升稳定性和可靠性。\n\n### v0.9.2\n* **批量自动整理功能** - 新增批量自动整理功能。现在您可以选择多个模型，并根据当前的路径模板设置自动整理它们，从而实现更高效的管理。\n* **Bug 修复** - 修复了若干 bug，以提升稳定性和可靠性。\n\n### v0.9.1\n* **增强的批量操作** - 改进了批量操作功能，新增套索选择和批量操作上下文菜单，提供更直观、类似桌面应用的用户体验。\n* **新增批量操作** - 新增批量为多个模型添加标签和设置基础模型的功能。\n\n### v0.9.0\n* **UI 全面升级，提升导航体验** - 将顶部的扁平文件夹标签替换为全新的文件夹侧边栏和面包屑导航系统，使文件夹浏览与选择更加直观便捷。\n* **双模式文件夹侧边栏** - 新增的文件夹侧边栏提供两种显示模式：“列表模式”模拟经典文件夹视图，“树状模式”则以层级结构呈现文件夹，方便用户轻松浏览嵌套目录。\n* **国际化支持** - 引入多语言支持，现已支持英语、简体中文、繁体中文、西班牙语、日语、韩语、法语、俄语和德语。欢迎母语人士反馈，帮助优化翻译质量。\n* **自动文件名冲突解决** - 实现文件自动重命名功能（`原名 + 短哈希值`），避免下载或移动模型时出现文件名冲突。\n* **性能优化与错误修复** - 进行了多项性能改进和错误修复，确保应用运行更稳定、响应更迅速。\n\n[查看更新历史](.\u002Fupdate_logs.md)\n\n---\n\n## **⚠ 重要提示**：要使用 CivitAI 下载功能，您需要：\n\n1. 在您的个人资料设置中获取 CivitAI API 密钥。\n2. 将其添加到 LoRA 管理器的设置页面。\n3. 保存设置。\n\n---\n\n## 核心功能\n\n- 🚀 **高性能**\n  - 快速加载和浏览模型\n  - 大型收藏集也能流畅滚动\n  \n- 🌐 **丰富的模型集成**\n  - 直接从 CivitAI 下载\n  - 支持预览图片和视频\n  - 显示模型描述及版本选择\n  - 一目了然地查看触发词\n  - 一键将预设值整合到工作流中\n  \n- 🔄 **检查点管理**\n  - 扫描并整理检查点模型\n  - 对收藏进行筛选和搜索\n  - 查看和编辑元数据\n  - 清理磁盘空间，高效管理存储\n  \n- 🧩 **LoRA 配方**\n  - 保存并分享喜爱的 LoRA 组合\n  - 记录生成参数以便日后参考\n  - 快速应用于工作流\n  - 支持导入导出功能，便于社区共享\n  \n- 💻 **用户友好**\n  - 可通过 ComfyUI 菜单一键访问\n  - 提供上下文菜单，快速执行操作\n  - 支持自定义备注和使用提示\n  - 多文件夹支持\n  - 可配置成人内容模糊阈值（`PG13` \u002F `R` \u002F `X` \u002F `XXX`，默认为 `R+`）\n    - 例如：将阈值设置为 `PG13` 时，在启用模糊功能的情况下，`PG13`、`R`、`X` 和 `XXX` 的预览都会被模糊处理\n  - 初始化过程中提供可视化进度指示\n\n---\n\n## 安装方法\n\n### 方法一：**ComfyUI 管理器**（推荐给 ComfyUI 用户）\n\n1. 打开 **ComfyUI**。\n2. 进入 **管理器 > 自定义节点管理器**。\n3. 搜索 `lora-manager`。\n4. 点击 **安装**。\n\n### 方法二：**便携式独立版**（无需 ComfyUI）\n\n1. 下载 [便携式压缩包](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Freleases\u002Fdownload\u002Fv1.0.0\u002Flora_manager_portable.7z)。\n2. 将提供的 `settings.json.example` 文件复制一份，重命名为 `settings.json`，放置于 `comfyui-lora-manager` 文件夹中。\n3. 编辑新创建的 `settings.json` 文件，填入正确的模型文件夹路径和 CivitAI API 密钥。\n   - 如果希望配置信息保留在仓库文件夹内而非用户设置目录，请将 `\"use_portable_settings\": true` 设置为真。\n4. 运行 `run.bat`。\n   - 如需更改启动端口，可编辑 `run.bat` 并修改参数（如 `--port 9001`）。\n\n### 方法三：**手动安装**\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager.git\ncd ComfyUI-Lora-Manager\npip install -r requirements.txt\n```\n\n## 使用说明\n\n1. 您可以通过两种方式访问 LoRA 管理器：\n   - 点击 ComfyUI 菜单中的“启动 LoRA 管理器”按钮。\n   - 直接访问 http:\u002F\u002Flocalhost:8188\u002Floras。\n2. 在界面中，您可以：\n   - 浏览并整理您的 LoRA 模型。\n   - 直接从 CivitAI 下载模型。\n   - 自动获取或手动设置预览图片。\n   - 查看并复制每个 LoRA 对应的触发词。\n   - 添加个人备注和使用建议。\n3. 若要在工作流中使用 LoRA：\n   - 将“Lora 加载器 (LoraManager)”节点添加到您的工作流中。\n   - 在管理器界面中选择一个 LoRA。\n   - 点击“复制”按钮或右键菜单中的“复制 LoRA 语法”选项。\n   - 将内容粘贴到 Lora 加载器节点的文本输入框中。\n   - 该节点会自动应用预设强度和触发词。\n\n### TextLM \u002F PromptLM 中的通配符功能\n\n`Text (LoraManager)` 和 `Prompt (LoraManager)` 支持 `\u002Fwildcard` 自动补全以及运行时通配符扩展。\n\n- 通配符文件存放在 `{settings 文件夹}\u002Fwildcards\u002F` 目录下。\n- 当您输入 `\u002Fwildcard` 且尚未创建任何通配符文件时，自动补全下拉菜单会显示确切的文件夹路径，并允许您直接打开该文件夹。\n- 支持的格式：`.txt`、`.yaml`、`.yml`、`.json`。\n\n格式规则：\n\n- `wildcards\u002Fanimals\u002Fcat.txt` 将变为 `__animals\u002Fcat__`。\n- `.txt` 文件每行仅包含一个选项。\n- YAML \u002F JSON 文件使用嵌套键，最终键值为字符串数组。\n\n示例：\n\n```txt\n# wildcards\u002Fcolor.txt\nred\nblue\ngreen\n```\n\n使用时写作 `__color__`。\n\n```yaml\n# wildcards\u002Fcolors.yaml\npalette:\n  warm:\n    - red\n    - orange\n```\n\n使用时写作 `__palette\u002Fwarm__`。\n\n### Save Image 节点的文件名格式模板\n\nSave Image 节点支持使用模式代码动态生成文件名。您可以按照以下格式模板自定义图像的命名方式：\n\n#### 可用的模式代码\n\n- `%seed%` - 插入生成种子编号。\n- `%width%` - 插入图像宽度。\n- `%height%` - 插入图像高度。\n- `%pprompt:N%` - 插入正面提示词（限制 N 个字符）。\n- `%nprompt:N%` - 插入负面提示词（限制 N 个字符）。\n- `%model:N%` - 插入模型\u002F检查点名称（限制 N 个字符）。\n- `%date%` - 插入当前日期时间，格式为“yyyyMMddhhmmss”。\n- `%date:FORMAT%` - 插入自定义格式的日期，其中：\n  - `yyyy` 表示 4 位年份。\n  - `yy` 表示 2 位年份。\n  - `MM` 表示 2 位月份。\n  - `dd` 表示 2 位日期。\n  - `hh` 表示 2 位小时。\n  - `mm` 表示 2 位分钟。\n  - `ss` 表示 2 位秒。\n\n#### 示例\n\n- `image_%seed%` → `image_1234567890`\n- `gen_%width%x%height%` → `gen_512x768`\n- `%model:10%_%seed%` → `dreamshape_1234567890`\n- `%date:yyyy-MM-dd%` → `2025-04-28`\n- `%pprompt:20%_%seed%` → `beautiful landscape_1234567890`\n- `%model%_%date:yyMMdd%_%seed%` → `dreamshaper_v8_250428_1234567890`\n\n您可以组合多种模式，为生成的图像创建详细且条理清晰的文件名。\n\n### 独立模式\n\n现在您可以将 LoRA Manager 从 ComfyUI 中独立运行：\n\n1. **对于 ComfyUI 用户**：\n   - 至少启动一次带有 LoRA Manager 的 ComfyUI，以便在用户设置文件夹中的 `settings.json` 文件中初始化必要的路径信息（参见上述路径）。\n   - 确保已安装依赖项：`pip install -r requirements.txt`\n   - 从您的 ComfyUI 根目录运行：\n     ```bash\n     python custom_nodes\\comfyui-lora-manager\\standalone.py\n     ```\n   - 您可以通过以下地址访问界面：`http:\u002F\u002Flocalhost:8188\u002Floras`\n   - 您也可以通过参数指定不同的主机或端口：\n     ```bash\n     python custom_nodes\\comfyui-lora-manager\\standalone.py --host 127.0.0.1 --port 9000\n     ```\n\n2. **对于非 ComfyUI 用户**：\n   - 复制提供的 `settings.json.example` 文件，创建一个名为 `settings.json` 的新文件。仅需更新 API 密钥、可选语言和文件夹路径——库注册表将在 LoRA Manager 启动时自动创建。\n   - 编辑 `settings.json`，填入正确的模型文件夹路径和 CivitAI API 密钥（您可以在准备好配置之前保留默认值）。\n   - 如果您希望 LoRA Manager 读写项目目录下的 `settings.json` 文件，可以启用便携模式，将 `\"use_portable_settings\": true` 设置为真。\n   - 安装所需依赖项：`pip install -r requirements.txt`\n   - 运行独立模式：\n     ```bash\n     python standalone.py\n     ```\n   - 您可以通过浏览器访问界面：`http:\u002F\u002Flocalhost:8188\u002Floras`\n\n   > **注意**：首次启动此版本时，现有安装会自动将旧版 `settings.json` 从插件文件夹迁移到用户设置目录。\n\n此独立模式提供了一种轻量级的选项，用于管理您的模型和配方集合，而无需运行完整的 ComfyUI 环境，因此即使主要使用其他稳定扩散界面的用户也会觉得它非常有用。\n\n## 测试与覆盖率\n\n### 后端\n\n安装开发依赖项，并运行带有覆盖率报告的 pytest：\n\n```bash\npip install -r requirements-dev.txt\nCOVERAGE_FILE=coverage\u002Fbackend\u002F.coverage pytest \\\n  --cov=py \\\n  --cov=standalone \\\n  --cov-report=term-missing \\\n  --cov-report=html:coverage\u002Fbackend\u002Fhtml \\\n  --cov-report=xml:coverage\u002Fbackend\u002Fcoverage.xml \\\n  --cov-report=json:coverage\u002Fbackend\u002Fcoverage.json\n```\n\nHTML、XML 和 JSON 报告文件将存储在 `coverage\u002Fbackend\u002F` 目录下，方便您在本地或通过 CI 工作流检查代码热点。\n\n### 前端\n\n运行 Vitest 覆盖率测试套件，以分析组件热点：\n\n```bash\nnpm run test:coverage\n```\n\n---\n\n## 文档\n\n- **[metadata.json 模式文档](docs\u002Fmetadata-json-schema.md)** — `.metadata.json` 附带文件格式的完整参考，包括所有字段、类型及 LoRA、检查点和嵌入模型的示例。\n\n---\n\n## 贡献\n\n感谢您对参与 ComfyUI LoRA Manager 开发的兴趣！由于该项目目前仍处于早期阶段，正处于快速开发和重构之中，我们暂时不接受拉取请求。\n\n不过，您的反馈和想法对我们来说极其重要：\n- 如发现任何问题，请随时提交问题\n- 通过 GitHub 问题提出功能需求\n- 分享您的改进建议\n\n我们感谢您的理解，并期待在项目架构稳定后能够接受代码贡献。\n\n---\n\n## 致谢\n\n本项目受到其他优秀 ComfyUI 扩展的启发，并从中受益：\n\n- [ComfyUI-QwenImageLoraLoader](https:\u002F\u002Fgithub.com\u002Fussoewwin\u002FComfyUI-QwenImageLoraLoader) — 提供实验性的 Nunchaku Qwen-Image LoRA 支持\n- [ComfyUI-SaveImageWithMetaData](https:\u002F\u002Fgithub.com\u002Fnkchocoai\u002FComfyUI-SaveImageWithMetaData) — 提供图像元数据功能\n- [rgthree-comfy](https:\u002F\u002Fgithub.com\u002Frgthree\u002Frgthree-comfy) — 提供 LoRA 加载功能\n\n---\n\n## ☕ 支持\n\n如果您觉得本项目有所帮助，请考虑支持其开发：\n\n[![ko-fi](https:\u002F\u002Fko-fi.com\u002Fimg\u002Fgithubbutton_sm.svg)](https:\u002F\u002Fko-fi.com\u002Fpixelpawsai)\n\n[![Patreon](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBecome%20a%20Patron-F96854.svg?style=for-the-badge&logo=patreon&logoColor=white)](https:\u002F\u002Fpatreon.com\u002FPixelPawsAI)\n\n微信：[点击查看二维码](https:\u002F\u002Fraw.githubusercontent.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Fmain\u002Fstatic\u002Fimages\u002Fwechat-qr.webp)\n\n## 💬 社区\n\n加入我们的 Discord 社区，获取支持、参与讨论并了解最新动态：\n[Discord 服务器](https:\u002F\u002Fdiscord.gg\u002FvcqNrWVFvM)\n\n---\n## 星标历史\n\n[![星标历史图表](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_readme_cb3548a2d67b.png)](https:\u002F\u002Fstar-history.com\u002F#willmiao\u002FComfyUI-Lora-Manager&Date)","# ComfyUI LoRA Manager 快速上手指南\n\nComfyUI LoRA Manager 是一款专为 ComfyUI 设计的强大工具，旨在简化 LoRA 模型的组织、下载和应用流程。它提供食谱管理、检查点整理以及一键工作流集成等功能，让模型管理工作更加高效流畅。\n\n## 环境准备\n\n在开始之前，请确保满足以下系统要求：\n\n*   **操作系统**: Windows, macOS 或 Linux\n*   **核心依赖**:\n    *   已安装并配置好 **ComfyUI** (建议最新版本)\n    *   **Python 3.10+** (与 ComfyUI 环境一致)\n    *   **Git** (用于克隆插件仓库)\n*   **网络环境**: 能够访问 GitHub 和 Civitai (若需下载模型)。国内用户若连接缓慢，建议配置全局代理或使用加速工具。\n\n## 安装步骤\n\n本工具作为 ComfyUI 的自定义节点插件安装。推荐使用 **ComfyUI Manager** 进行安装，也可手动安装。\n\n### 方法一：通过 ComfyUI Manager 安装（推荐）\n\n1.  启动 ComfyUI，点击右侧菜单中的 **Manager** 按钮。\n2.  选择 **Install Custom Nodes**。\n3.  在搜索框中输入 `LoRA Manager` 或 `ComfyUI-Lora-Manager`。\n4.  找到由 `willmiao` 开发的插件，点击 **Install**。\n5.  安装完成后，**重启 ComfyUI** 以生效。\n\n### 方法二：手动安装\n\n如果未安装 Manager，可通过命令行手动安装：\n\n1.  进入 ComfyUI 的自定义节点目录：\n    ```bash\n    cd path\u002Fto\u002FComfyUI\u002Fcustom_nodes\n    ```\n    *(请将 `path\u002Fto\u002FComfyUI` 替换为你实际的 ComfyUI 安装路径)*\n\n2.  克隆仓库：\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager.git\n    ```\n    > **国内加速提示**: 若克隆速度慢，可使用镜像源：\n    > `git clone https:\u002F\u002Fghp.ci\u002Fhttps:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager.git`\n\n3.  安装依赖（通常 ComfyUI 启动时会自动处理，若报错可手动执行）：\n    ```bash\n    cd ComfyUI-Lora-Manager\n    pip install -r requirements.txt\n    ```\n\n4.  重启 ComfyUI。\n\n## 基本使用\n\n安装并重启后，你可以通过以下方式开始使用：\n\n### 1. 访问管理界面\n启动 ComfyUI 后，在浏览器中打开以下地址即可进入 LoRA Manager 专属界面：\n```text\nhttp:\u002F\u002Flocalhost:8188\u002Floras\n```\n*(如果你的 ComfyUI 运行在其他端口，请相应修改端口号)*\n\n### 2. 浏览与管理模型\n*   **模型列表**: 界面将自动扫描并展示本地 `models\u002Floras` 目录下的所有 LoRA 模型。\n*   **一键下载**: 配合浏览器扩展（见下文），可直接从 Civitai 页面点击下载按钮，模型将自动保存至正确目录。\n*   **食谱管理**: 在界面中查看模型的生成参数（Recipe），支持一键应用到当前工作流。\n\n### 3. 一键工作流集成\n这是该工具的核心功能之一：\n1.  在 LoRA Manager 网页界面中找到想要的模型。\n2.  点击 **Integrate** 或类似的一键添加按钮。\n3.  返回 ComfyUI 主界面，你会发现对应的 **Load LoRA** 节点已自动添加到工作流中，并预填了模型路径和参数，无需手动拖拽和配置。\n\n### 4. (可选) 安装浏览器扩展\n为了获得最佳的“一键下载”体验，建议安装配套的浏览器扩展：\n*   **Chrome\u002FEdge**: [Chrome Web Store 链接](https:\u002F\u002Fchromewebstore.google.com\u002Fdetail\u002Flm-civitai-extension\u002Fcapigligggeijgmocnaflanlbghnamgm)\n*   **Firefox**: [Firefox 扩展下载](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002Flm-civitai-extension-firefox\u002Freleases\u002Flatest\u002Fdownload\u002Fextension.xpi)\n\n安装扩展后，访问 Civitai 模型页时，你会看到额外的状态指示（是否已下载）和快速下载按钮。","一位专注于角色设计的数字艺术家正在 ComfyUI 中批量生成不同风格的二次元角色图，需要频繁切换和测试数十个不同的 LoRA 模型以寻找最佳效果。\n\n### 没有 ComfyUI-Lora-Manager 时\n- **模型管理混乱**：本地存放着上百个未分类的 LoRA 文件，仅靠文件名难以回忆具体风格，每次尝试新模型都要在文件夹中盲目翻找。\n- **工作流打断严重**：在 Civitai 浏览到新模型后，需手动复制链接、打开下载器、等待完成，再回到 ComfyUI 刷新节点并重新连接，流程繁琐且易出错。\n- **参数调试低效**：无法直观预览模型效果或查看元数据，往往需要构建完整工作流并运行一次后才能知道该 LoRA 是否适用，极大浪费算力与时间。\n- **协作共享困难**：想要复现某个特定角色的生成效果时，难以快速记录并分享包含具体 LoRA 版本和参数配置的完整“配方”。\n\n### 使用 ComfyUI-Lora-Manager 后\n- **可视化高效组织**：通过内置的 Web 界面直接预览所有 LoRA 缩略图与元数据，支持标签分类和搜索，瞬间定位所需模型。\n- **一键集成工作流**：配合浏览器扩展，在 Civitai 页面即可检测已拥模型并一键下载，随后在 ComfyUI 中点击即可自动将 LoRA 节点插入当前工作流。\n- **即时验证与复用**：无需运行完整生成即可预览模型信息，利用“配方管理”功能保存成功的工作流组合，随时一键调用或分享给团队成员。\n- **闭环体验流畅**：从发现、下载、管理到应用，全流程在统一生态内完成，让创作者能专注于创意调整而非文件运维。\n\nComfyUI-Lora-Manager 将原本碎片化、高摩擦的模型试错过程，转化为流畅可视化的创意探索之旅，显著提升了艺术创作的迭代效率。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fwillmiao_ComfyUI-Lora-Manager_6939594b.png","willmiao","pixelpaws","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fwillmiao_c4ef2bda.jpg",null,"Xi'an, China","willmiao.dev@gmail.com","https:\u002F\u002Fgithub.com\u002Fwillmiao",[81,85,89,93,97,101,105],{"name":82,"color":83,"percentage":84},"Python","#3572A5",48,{"name":86,"color":87,"percentage":88},"JavaScript","#f1e05a",39.4,{"name":90,"color":91,"percentage":92},"CSS","#663399",5.1,{"name":94,"color":95,"percentage":96},"HTML","#e34c26",3.1,{"name":98,"color":99,"percentage":100},"Vue","#41b883",2.5,{"name":102,"color":103,"percentage":104},"TypeScript","#3178c6",1.8,{"name":106,"color":107,"percentage":108},"Shell","#89e051",0,1074,103,"2026-04-16T04:08:25","GPL-3.0","未说明",{"notes":115,"python":113,"dependencies":116},"该工具是 ComfyUI 的插件（自定义节点），并非独立运行的应用程序，因此其运行环境完全依赖于宿主 ComfyUI 的配置。它主要提供 LoRA 模型的管理、下载和组织功能，界面访问地址为 http:\u002F\u002Flocalhost:8188\u002Floras。此外，项目还提供了一个配套的浏览器扩展（支持 Chrome 和 Firefox），用于在 Civitai 网站上直接查看已拥有的模型并一键下载。",[56],[15,27],"2026-03-27T02:49:30.150509","2026-04-20T16:33:34.286427",[],[122,127,132,137,142,147,152,157,162,167,172,177,182,187,192,197,202,207,212,217],{"id":123,"version":124,"summary_zh":125,"released_at":126},289050,"v1.0.3","### v1.0.3\n\n* **自定义配方存储路径** - 新增支持为配方配置自定义存储路径，并提供迁移功能，可在更改路径时自动转移现有配方数据。\n* **LM 文本\u002F提示节点的通配符支持** - LM 的 `Text` 节点和 `Prompt` 节点现支持新的 `\u002Fwildcard` 命令，可在运行时展开通配符，并支持动态提示语法，以实现更灵活的提示构建。\n* **系统诊断（“医生”）** - 新增诊断功能，帮助更清晰地发现环境和设置方面的问题。\n* **用户状态备份支持** - 新增用户状态备份功能，配套提供相关界面，并在设置中明确提示备份范围。\n* **已下载状态可见性** - 优化了已下载状态的交互体验，使先前下载过的模型版本更加易于识别。\n* **自动补全性能优化** - 修复了自动补全的性能问题，降低了标签搜索的开销，提升了响应速度。","2026-04-15T15:13:43",{"id":128,"version":129,"summary_zh":130,"released_at":131},289051,"v1.0.2","### v1.0.2\n\n* **模型下载历史追踪** - LoRA 管理器现在会保存已下载模型版本的历史记录，从而能够识别某个版本是否曾经被下载过，即使该版本当前已不在您的库中。\n* **跳过已下载过的模型版本** - 新增了一个设置项“跳过已下载过的模型版本”，以帮助避免重复下载您之前已经下载过的模型版本。\n* **LoRA 堆栈合并节点触发词修复** - 修复了一个问题：来自 `LORA_STACK` 输入的触发词更新无法正确传递到 LoRA 堆栈合并节点。\n* **CivitAI 示例图片兼容性改进** - 改进了对 CivitAI CDN 子域名的支持，使示例图片加载更加稳定可靠。","2026-04-06T12:16:56",{"id":133,"version":134,"summary_zh":135,"released_at":136},289052,"v1.0.1","### v1.0.1\n\n* **批量导入配方** - 支持同时从多个 URL 或目录导入配方，并采用优化的并发机制。\n* **批量下载缺失 LoRA** - 新增配方批量操作：选择多个配方，一次性下载所有未下载的 LoRA 文件。\n* **仅导入配方选项** - 只保存配方元数据，不下载缺失的 LoRA，方便您先收藏感兴趣的配方，待需要时再下载依赖项。\n* **可编辑配方提示词** - 直接在配方详情弹窗中编辑提示词。\n* **检查点加载 LM 节点** - 行为与 ComfyUI 内置的“加载检查点”节点一致，额外支持从“额外文件夹路径”加载检查点。\n* **UNet 加载 LM 节点** - 行为与 ComfyUI 内置的“加载扩散模型”节点一致，支持从“额外文件夹路径”加载，并兼容 GGUF 格式（需配合 ComfyUI-GGUF 自定义节点）。\n* **LoRA 堆栈合并节点** - 将两个 LoRA 堆栈合并为一个。例如：分别为角色和风格 LoRA 使用独立的随机化器，然后再合并应用。\n* **LoRA 池正则筛选** - 通过自定义正则表达式规则（包含\u002F排除），筛选进入池中的 LoRA。\n* **动态基础模型类型** - 现在会从 Civitai API 动态获取基础模型类型，确保与最新可用模型保持同步。\n* **提示词自动补全增强** - 支持 Tab 键确认、可配置行为，以及改进的多词标签匹配功能。\n* **基础模型下载排除** - 在仅需特定类型模型时，可将某些基础模型从下载任务中排除。\n* **成人内容模糊阈值设置** - 可为成人内容预览配置模糊等级（`PG13` \u002F `R` \u002F `X` \u002F `XXX`，默认为 `R+`）。\n* **实验性：Nunchaku Qwen LoRA 支持** - 实验性支持加载并应用 LoRA 到 Nunchaku 量化版 Qwen-Image 模型。\n* **Bug 修复与 UX 优化** - 多项修复，提升工作流流畅度。","2026-04-02T03:56:27",{"id":138,"version":139,"summary_zh":140,"released_at":141},289053,"v1.0.0","### v1.0.0\n* **支持额外的文件夹路径** - 新增了对 LoRA 管理器专用的额外模型根路径的支持。这使得用户可以从 ComfyUI 标准文件夹之外的其他位置加载 LoRA 模型，从而在处理大型模型库时避免性能问题。\n* **设置界面全面重构** - 重新设计了设置界面，布局更加清晰有序，方便用户快速找到并配置应用的各项设置。\n* **惰性哈希计算** - 对大型模型文件（检查点和扩散模型）实现了惰性哈希计算。哈希值仅在绝对必要时才会被计算，从而最大限度地减少不必要的磁盘 I\u002FO 操作，并显著加快应用的初始化速度。\n* **里程碑与支持者表彰** - 更新了“支持者”窗口，在迎来 v1.0.0 这一里程碑之际，向所有项目支持者表达感谢。衷心感谢社区一直以来的支持！\n* **Bug 修复与用户体验优化** - 修复了多个 Bug，并对用户体验进行了多项优化，以提供更流畅的工作流程。","2026-03-03T14:06:14",{"id":143,"version":144,"summary_zh":145,"released_at":146},289054,"v0.9.16","### v0.9.16\n* **重复检测增强** - 模型重复模式现在会尊重筛选器配置，从而更轻松地在特定筛选结果中查找重复组。\n* **标签逻辑切换** - 在筛选器面板中为包含标签筛选添加了 OR\u002FAND 切换开关，提供更灵活的基于标签的模型搜索功能。\n* **元数据刷新跳过路径** - 新增设置，可将特定路径排除在元数据刷新操作之外。从远程源获取元数据时，这些路径下的模型将被跳过。\n* **提示节点中的动态触发词** - 提示节点现在支持动态数量的触发词输入，以实现更高的灵活性。\n* **抢先体验更新** - 模型更新现会显示抢先体验信息，并新增一项设置，允许用户根据需要忽略抢先体验更新。\n* **LM Civitai 扩展集成** - 新增与 LM Civitai 扩展的集成。点击模型更新中的下载按钮后，下载任务将自动发送至该扩展的下载队列，实现无缝的一键下载。","2026-02-22T13:08:36",{"id":148,"version":149,"summary_zh":150,"released_at":151},289055,"v0.9.15","### v0.9.15\n* **滤镜预设** - 将滤镜组合保存为预设，以便快速切换和重新应用。\n* **错误修复** - 修复了多个 bug，提升了稳定性。","2026-02-04T01:13:29",{"id":153,"version":154,"summary_zh":155,"released_at":156},289056,"v0.9.14","### v0.9.14\n* **LoRA 循环节点** - 新增 LoRA 循环节点，支持按指定顺序循环加载多个 LoRA 模型，并可设置重复次数及暂停循环功能。具体用法请参考新的“Lora Cycler”模板工作流。\n* **增强的提示词节点与标签自动补全** - 提示词节点现已支持基于 Danbooru 和 e621 标签合并后的全面标签自动补全功能，包含标签搜索和自动补全。新增命令系统，可通过 `\u002Fchar` 或 `\u002Fartist` 等快捷指令进行特定分类的标签搜索；同时提供 `\u002Fac` 和 `\u002Fnoac` 命令，用于快速开启或关闭自动补全功能。详细使用技巧请参阅 ComfyUI -> 模板 -> ComfyUI-Lora-Manager 中的“Lora Manager Basic”模板工作流。\n* **Bug 修复与稳定性提升** - 修复了多个 Bug，并进一步提升了整体稳定性。","2026-02-02T15:19:13",{"id":158,"version":159,"summary_zh":160,"released_at":161},289057,"v0.9.13","### v0.9.13 修复更新至版本 0.9.12  \n* **LoRA 随机化系统** - 引入了全面的 LoRA 随机化系统，包含 LoRA 池和 LoRA 随机化节点，用于灵活、动态的生成工作流。  \n* **LoRA 随机化模板** - 请参考全新的“LoRA 随机化”模板工作流，其中提供了灵活随机化模式、锁定与复用选项等详细示例。  \n* **配方文件夹** - 在配方页面引入了文件夹系统，用户可以像管理模型一样自由地组织配方。  \n* **配方批量操作** - 新增批量模式支持，可通过点击拖选、拖拽至文件夹以及 Ctrl+A（全选）等直观操作，对选定的配方进行批量移动、删除及设置基础模型。  \n* **提示词搜索与排序** - 可按提示词内容搜索配方，并按配方名称、导入日期或 LoRA 数量进行排序，以提升浏览体验。  \n* **配方收藏** - 将特定配方标记为收藏，方便快速访问。  \n* **视频配方支持** - 开启了对视频配方的支持（可通过 LM 扩展或 URL 导入；暂不支持直接导入视频文件）。  \n* **性能优化** - 修复了性能问题，显著提升了启动和加载速度。首次扫描完成后，无论收藏规模如何，后续加载都将实现秒开。  \n* **ComfyUI Nodes 2.0 支持** - 提供对 ComfyUI Nodes 2.0 的基础支持。","2026-01-21T13:23:24",{"id":163,"version":164,"summary_zh":165,"released_at":166},289058,"v0.9.12","### v0.9.12 重大更新\n* **LoRA 随机化系统** - 引入了全面的 LoRA 随机化系统，包含 LoRA 池和 LoRA 随机化节点，用于实现灵活且动态的生成工作流。\n* **LoRA 随机化模板** - 请参考全新的“LoRA 随机化”模板工作流，其中提供了灵活随机化模式、锁定与复用选项等特性的详细示例。\n* **配方文件夹** - 在配方页面引入了文件夹系统，用户可以像管理模型一样自由地组织配方。\n* **配方批量操作** - 新增批量模式支持，可通过直观的操作方式（如点击拖选、拖拽至文件夹、Ctrl+A 全选）对选定的配方进行批量移动、删除以及设置基础模型。\n* **提示词搜索与排序** - 可按提示词内容搜索配方，并按配方名称、导入日期或 LoRA 数量进行排序，以提升浏览体验。\n* **配方收藏** - 将特定配方标记为收藏，方便快速访问。\n* **视频配方支持** - 开启了对视频配方的支持（可通过 LM 扩展或 URL 导入；暂不支持直接导入视频文件）。\n* **性能优化** - 修复了多项性能问题，显著提升了启动和加载速度。首次扫描完成后，无论收藏规模大小，后续加载都将瞬间完成。\n* **ComfyUI Nodes 2.0 支持** - 提供对 ComfyUI Nodes 2.0 的基础支持。","2026-01-19T09:40:14",{"id":168,"version":169,"summary_zh":170,"released_at":171},289059,"v0.9.11","* **Bug 修复** - 修复了多个 bug，以提升稳定性和可靠性。","2025-11-29T09:47:50",{"id":173,"version":174,"summary_zh":175,"released_at":176},289060,"v0.9.10","### v0.9.10\r\n* **Smarter Update Matching** - Users can now choose to check and group updates by matching base model only or with no base-model constraint; version lists also support toggling between same-base versions or all versions.\r\n* **Flexible Tag Filtering** - The filter panel now supports tag exclusion: click a tag to include, click again to exclude, and click a third time to clear, enabling stronger and more flexible tag filters.\r\n* **License Visibility & Controls** - Model detail headers and ComfyUI preview popups now show Civitai license icons. The filter panel gains license include\u002Fexclude options, and a new global context menu action, \"Refresh license metadata,\" fetches missing license data.\r\n* **Recipe Improvements** - Recipes now allow importing with zero LoRAs, and recipe detail pages show the related checkpoint for easier reference.\r\n* **Better ZIP Downloads** - When downloading models packaged in ZIPs, model files are extracted into the target model folder; ZIPs containing multiple model files (e.g., WanVideo high\u002Flow LoRA pairs) are added as separate models.\r\n* **Template Workflow Update** - Refreshed the \"Illustrious Pony Example\" template workflow with usage guidance for each LoRA Manager node.\r\n* **Bug Fixes & Stability** - General fixes and stability improvements.","2025-11-24T03:15:46",{"id":178,"version":179,"summary_zh":180,"released_at":181},289061,"v0.9.9","### v0.9.9\r\n* **Check for Updates Feature** - Users can now check for updates for all models or selected models in bulk mode. Models with available updates will display an \"update available\" badge on their model card, and users can filter to show only models with updates.\r\n* **Model Versions Management** - Added a new Versions tab in the model modal that centralizes all versions of a model, providing download, delete, and ignore update functions.\r\n* **Send Checkpoint to ComfyUI** - Users can now click the send button on a checkpoint card to send the checkpoint directly to the current workflow's checkpoint or diffusion model loader node in ComfyUI.\r\n* **Customizable Model Card Display** - Added a new setting that allows users to choose whether to display the model name or filename on model cards.\r\n* **New Path Template Placeholders** - Added new path template placeholders: `{model_name}` and `{version_name}` for more flexible organization.\r\n* **ComfyUI Auto Path Correction Setting** - Added a new setting within ComfyUI to enable or disable the auto path correction feature.\r\n","2025-11-03T14:41:31",{"id":183,"version":184,"summary_zh":185,"released_at":186},289062,"v0.9.8","### v0.9.8\r\n* **Full CivArchive API Support** - Added complete support for the CivArchive API as a fallback metadata source beyond Civitai API. Models deleted from Civitai can now still retrieve metadata through the CivArchive API.\r\n* **Download Models from CivArchive** - Added support for downloading models directly from CivArchive, similar to downloading from Civitai. Simply click the Download button and paste the model URL to download the corresponding model.\r\n* **Custom Priority Tags** - Introduced Custom Priority Tags feature, allowing users to define custom priority tags. These tags will appear as suggestions when editing tags or during auto organization\u002Fdownload using default paths, providing more precise and controlled folder organization. [Guide](https:\u002F\u002Fgithub.com\u002Fwillmiao\u002FComfyUI-Lora-Manager\u002Fwiki\u002FPriority-Tags-Configuration-Guide)\r\n* **Drag and Drop Tag Reordering** - Added drag and drop functionality to reorder tags in the tags edit mode for improved usability.\r\n* **Download Control in Example Images Panel** - Added stop control in the Download Example Images Panel for better download management.\r\n* **Prompt (LoraManager) Node with Autocomplete** - Added new Prompt (LoraManager) node with autocomplete feature for adding embeddings.\r\n* **Lora Manager Nodes in Subgraphs** - Lora Manager nodes now support being placed within subgraphs for more flexible workflow organization.","2025-10-15T11:48:16",{"id":188,"version":189,"summary_zh":190,"released_at":191},289063,"v0.9.7","### v0.9.7\r\n* **Critical Performance Optimization** - Introduced persistent model cache that dramatically accelerates initialization after startup and significantly reduces Python backend memory footprint for improved application performance.\r\n* **Cross-Browser Settings Synchronization** - Migrated nearly all settings to the backend, ensuring your preferences sync automatically across all browsers for a seamless multi-browser experience.\r\n* **Protected User Settings Location** - Relocated user settings (settings.json) to the user config directory (accessible via the link icon in Settings), preventing accidental deletion during reinstalls or updates.\r\n* **Global Context Menu** - Added a new global context menu accessible by right-clicking on empty page areas, providing quick access to global operations with more features coming in future updates.\r\n* **Multi-Library Support** - Introduced support for managing multiple libraries, allowing you to easily switch between different model collections (advanced usage, documentation in progress).\r\n* **Bug Fixes & Stability Improvements** - Various bug fixes and enhancements for improved stability and reliability.","2025-10-09T14:16:54",{"id":193,"version":194,"summary_zh":195,"released_at":196},289064,"v0.9.6","### v0.9.6\r\n* **Critical Performance Optimization** - Introduced persistent model cache that dramatically accelerates initialization after startup and significantly reduces Python backend memory footprint for improved application performance.\r\n* **Cross-Browser Settings Synchronization** - Migrated nearly all settings to the backend, ensuring your preferences sync automatically across all browsers for a seamless multi-browser experience.\r\n* **Protected User Settings Location** - Relocated user settings (settings.json) to the user config directory (accessible via the link icon in Settings), preventing accidental deletion during reinstalls or updates.\r\n* **Global Context Menu** - Added a new global context menu accessible by right-clicking on empty page areas, providing quick access to global operations with more features coming in future updates.\r\n* **Multi-Library Support** - Introduced support for managing multiple libraries, allowing you to easily switch between different model collections (advanced usage, documentation in progress).\r\n* **Bug Fixes & Stability Improvements** - Various bug fixes and enhancements for improved stability and reliability.","2025-10-07T09:42:59",{"id":198,"version":199,"summary_zh":200,"released_at":201},289065,"v0.9.5","### v0.9.5\r\n* Addressed an unexpected metadata fetching failure caused by Civitai API issues, restoring normal functionality.","2025-09-24T01:22:38",{"id":203,"version":204,"summary_zh":205,"released_at":206},289066,"v0.9.4","### v0.9.4\r\n* **Bug Fixes** - Various bug fixes for improved stability and reliability.","2025-09-20T14:04:51",{"id":208,"version":209,"summary_zh":210,"released_at":211},289067,"v0.9.3","### v0.9.3\r\n* **Metadata Archive Database Support** - Added the ability to download and utilize a metadata archive database, enabling access to metadata for models that have been deleted from CivitAI.\r\n* **App-Level Proxy Settings** - Introduced support for configuring a global proxy within the application, making it easier to use the manager behind network restrictions.\r\n* **Bug Fixes** - Various bug fixes for improved stability and reliability.","2025-09-16T10:56:57",{"id":213,"version":214,"summary_zh":215,"released_at":216},289068,"v0.9.2","### v0.9.2\r\n* **Bulk Auto-Organization Action** - Added a new bulk auto-organization feature. You can now select multiple models and automatically organize them according to your current path template settings for streamlined management.\r\n* **Bug Fixes** - Addressed several bugs to improve stability and reliability.","2025-09-06T06:39:52",{"id":218,"version":219,"summary_zh":220,"released_at":221},289069,"v0.9.1","### v0.9.1\r\n* **Enhanced Bulk Operations** - Improved bulk operations with Marquee Selection and a bulk operation context menu, providing a more intuitive, desktop-application-like user experience.\r\n* **New Bulk Actions** - Added bulk operations for adding tags and setting base models to multiple models simultaneously.","2025-09-05T10:15:43"]