[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-71193":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":34,"readmeContent":35,"aiSummary":36,"trendingCount":16,"starSnapshotCount":16,"syncStatus":18,"lastSyncTime":37,"discoverSource":38},71193,"multidiffusion-upscaler-for-automatic1111","pkuliyi2015\u002Fmultidiffusion-upscaler-for-automatic1111","pkuliyi2015","Tiled Diffusion and VAE optimize, licensed under CC BY-NC-SA 4.0","",null,"Python",4999,346,49,123,0,1,2,4,3,29.62,"Other",false,"main",true,[27,28,29,30,31,32,33],"image-generation","large-image","multidiffusion","stable-diffusion","stable-diffusion-webui","stable-diffusion-webui-plugin","vramsaving","2026-06-12 02:02:49","# Tiled Diffusion & VAE extension for sd-webui\n\n[![CC BY-NC-SA 4.0][cc-by-nc-sa-shield]][cc-by-nc-sa]\n\nThis extension is licensed under [CC BY-NC-SA](https:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby-nc-sa\u002F4.0\u002F), everyone is FREE of charge to access, use, modify and redistribute with the same license.  \n**You cannot use versions after AOE 2023.3.28 for commercial sales (only refers to code of this repo, the derived artworks are NOT restricted).**\n\n由于部分无良商家销售WebUI，捆绑本插件做卖点收取智商税，本仓库的许可证已修改为 [CC BY-NC-SA](https:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby-nc-sa\u002F4.0\u002F)，任何人都可以自由获取、使用、修改、以相同协议重分发本插件。  \n**自许可证修改之日(AOE 2023.3.28)起，之后的版本禁止用于商业贩售 (不可贩售本仓库代码，但衍生的艺术创作内容物不受此限制)。**\n\nIf you like the project, please give me a star! ⭐\n\n[![ko-fi](https:\u002F\u002Fko-fi.com\u002Fimg\u002Fgithubbutton_sm.svg)](https:\u002F\u002Fko-fi.com\u002Fpkuliyi2015)\n\n****\n\n\nThe extension helps you to **generate or upscale large images (≥2K) with limited VRAM (≤6GB)** via the following techniques:\n\n- Reproduced SOTA Tiled Diffusion methods\n  - [Mixture of Diffusers](https:\u002F\u002Fgithub.com\u002Falbarji\u002Fmixture-of-diffusers)\n  - [MultiDiffusion](https:\u002F\u002Fmultidiffusion.github.io)\n  - [Demofusion](https:\u002F\u002Fgithub.com\u002FPRIS-CV\u002FDemoFusion)\n- Our original Tiled VAE method\n- My original Tiled Noise Inversion method\n\n\n### Features\n\n- Core\n  - [x] [Tiled VAE](#tiled-vae)\n  - [x] [Tiled Diffusion: txt2img generation for ultra-large image](#tiled-diff-txt2img)\n  - [x] [Tiled Diffusion: img2img upscaling for image detail enhancement](#tiled-diff-img2img)\n  - [x] [Regional Prompt Control](#region-prompt-control)\n  - [x] [Tiled Noise Inversion](#tiled-noise-inversion)\n- Advanced\n  - [x] [ControlNet support]()\n  - [x] [StableSR support](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fsd-webui-stablesr)\n  - [x] [SDXL support](experimental)\n  - [x] [Demofusion support]()\n\n👉 在 [wiki](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-upscaler-for-automatic1111\u002Fwiki) 页面查看详细的文档和样例，以及由 [@PotatoBananaApple](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-upscaler-for-automatic1111\u002Fdiscussions\u002F120) 制作的 [快速入门教程](https:\u002F\u002Fcivitai.com\u002Fmodels\u002F34726)\n👉 Find detailed documentation & examples at our [wiki](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-upscaler-for-automatic1111\u002Fwiki), and quickstart [Tutorial](https:\u002F\u002Fcivitai.com\u002Fmodels\u002F34726) by [@PotatoBananaApple](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-upscaler-for-automatic1111\u002Fdiscussions\u002F120) 🎉\n\n\n### Examples\n\n⚪ Txt2img: generating ultra-large images\n\n`prompt: masterpiece, best quality, highres, city skyline, night.`\n\n![panorama](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fcity_panorama.jpeg?raw=true)\n\n⚪ Img2img: upcaling for detail enhancement\n\n| original | x4 upscale |\n| :-: | :-: |\n| ![lowres](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Flowres.jpg?raw=true) | ![highres](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fhighres.jpeg?raw=true) |\n\n⚪ Regional Prompt Control\n\n| region setting | output1 | output2 |\n| :-: | :-: | :-: |\n| ![MultiCharacterRegions](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fmulticharacter.png?raw=true) | ![MultiCharacter](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fmulticharacter.jpeg?raw=true) | ![MultiCharacter](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fmulticharacter2.jpeg?raw=true) |\n| ![FullBodyRegions](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Ffullbody_regions.png?raw=true) | ![FullBody](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Ffullbody.jpeg?raw=true) | ![FullBody2](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Ffullbody2.jpeg?raw=true) |\n\n⚪ ControlNet support\n\n| original | with canny |\n| :-: | :-: |\n| ![Your Name](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fyourname_canny.jpeg?raw=true) | ![Your Name](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fyourname.jpeg?raw=true)\n\n|  | 重绘 “清明上河图” |\n| :-: | :-: |\n| original | ![ancient city](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fancient_city_origin_compressed.jpeg?raw=true) |\n| processed  | ![ancient city](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-img-demo\u002Fblob\u002Fmaster\u002Fancient_city_compressed.jpeg?raw=true) |\n\n⚪ DemoFusion support\n\n| original | x3 upscale |\n| :-: | :-: |\n| ![demo-example](https:\u002F\u002Fgithub.com\u002FJaylen-Lee\u002Fimage-demo\u002Fblob\u002Fmain\u002Fexample.png?raw=true) | ![demo-result](https:\u002F\u002Fgithub.com\u002FJaylen-Lee\u002Fimage-demo\u002Fblob\u002Fmain\u002F3.png?raw=true) |\n\n\n### License\n\nGreat thanks to all the contributors! 🎉🎉🎉  \nThis work is licensed under [Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License][cc-by-nc-sa].\n\n[![CC BY-NC-SA 4.0][cc-by-nc-sa-image]][cc-by-nc-sa]\n[![CC BY-NC-SA 4.0][cc-by-nc-sa-shield]][cc-by-nc-sa]\n\n[cc-by-nc-sa]: http:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby-nc-sa\u002F4.0\u002F\n[cc-by-nc-sa-image]: https:\u002F\u002Flicensebuttons.net\u002Fl\u002Fby-nc-sa\u002F4.0\u002F88x31.png\n[cc-by-nc-sa-shield]: https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-CC%20BY--NC--SA%204.0-lightgrey.svg\n","该项目是一个用于Stable Diffusion WebUI的扩展，旨在通过分块扩散和VAE优化技术生成或放大高分辨率图像（≥2K），尤其适用于显存资源有限（≤6GB）的情况。它集成了多种先进的分块扩散方法，如Mixture of Diffusers、MultiDiffusion和Demofusion，并引入了原创的分块VAE和噪声逆变方法，以提高图像生成质量和细节增强效果。此外，该插件还支持ControlNet、StableSR以及实验性的SDXL等功能，增强了其在复杂场景下的应用能力。此工具非常适合需要在有限硬件条件下处理大尺寸图片的用户，例如艺术家、设计师或是任何对高质量图像生成有需求的人士。","2026-06-11 03:36:31","high_star"]