{"id":18742,"date":"2024-08-27T10:18:51","date_gmt":"2024-08-27T02:18:51","guid":{"rendered":"https:\/\/www.1ai.net\/?p=18742"},"modified":"2024-08-27T10:19:00","modified_gmt":"2024-08-27T02:19:00","slug":"%e8%bd%bb%e6%9d%be%e4%b8%8a%e6%89%8b%ef%bc%8cflux%e5%ae%89%e8%a3%85%e6%95%99%e7%a8%8b%ef%bc%8c6g%e6%98%be%e5%ad%98%e5%8f%af%e7%94%a8","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/18742.html","title":{"rendered":"Easy to get started, Flux installation tutorial, 6G video memory available"},"content":{"rendered":"<div data-block=\"true\" data-editor=\"wyh\" data-offset-key=\"5n66j-0-0\">\n<div class=\"public-DraftStyleDefault-block public-DraftStyleDefault-ltr\" data-offset-key=\"5n66j-0-0\">\n<p>I have shared some experiences with you before.<a href=\"https:\/\/www.1ai.net\/en\/tag\/flux%e6%a8%a1%e5%9e%8b\" title=\"_Other Organiser\" target=\"_blank\" >Flux Model<\/a>way.<\/p>\n<p><a href=\"https:\/\/www.1ai.net\/en\/17942.html\/\">Flux [Basics]: Share some websites where you can experience the Flux.1 model online<\/a><\/p>\n<p><a href=\"https:\/\/www.1ai.net\/en\/18204.html\/\">Flux [Basics]: Local deployment and installation tutorial of ComfyUI Flux.1 workflow<\/a><\/p>\n<p>Today we will share how to deploy locally<a href=\"https:\/\/www.1ai.net\/en\/tag\/comfyui\" title=\"_Other Organiser\" target=\"_blank\" >ComfyUI<\/a> Flux.1 dev GGUF workflow.<\/p>\n<p><strong>1. First acquaintance with GGUF<\/strong><\/p>\n<p>GGUF stands for GPT-Generated Unified Format, a binary file format designed for large language models. It was proposed by Georgi Gerganov, the founder of llama.cpp, and is used to efficiently store and exchange pre-training results of large models. This format optimizes the data structure and encoding method, significantly improves the storage efficiency of model files, and ensures fast loading performance.<\/p>\n<p>On the official website of XLabs-AI on Github, GGUF is also the first choice for low-memory usage.<\/p>\n<blockquote>\n<ul>\n<li>https:\/\/github.com\/XLabs-AI\/x-flux-comfyui<\/li>\n<\/ul>\n<\/blockquote>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18743\" title=\"a171db98j00siuu3t000sd000oa007bm\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/a171db98j00siuu3t000sd000oa007bm.jpg\" alt=\"a171db98j00siuu3t000sd000oa007bm\" width=\"874\" height=\"263\" \/><\/p>\n<p>In addition, many friends have heard of or used another way to use low video memory: Flux.1 dev NF4, which is currently abandoned.<\/p>\n<blockquote>\n<ul>\n<li>https:\/\/github.com\/comfyanonymous\/ComfyUI_bitsandbytes_NF4<\/li>\n<\/ul>\n<\/blockquote>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18747\" title=\"1b74f188j00siuu3t001yd000u000j9m\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/1b74f188j00siuu3t001yd000u000j9m.jpg\" alt=\"1b74f188j00siuu3t001yd000u000j9m\" width=\"1080\" height=\"693\" \/><\/p>\n<p><strong>2. Flux.1 dev GGUF workflow construction<\/strong><\/p>\n<p>We will take the following typical ComfyUI Flux.1 dev GGUF workflow as an example to explain in detail (the network disk address can be obtained at the end of the workflow article).<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18744\" title=\"2da7d526j00siuu3t0013d000u000e8m\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/2da7d526j00siuu3t0013d000u000e8m.jpg\" alt=\"2da7d526j00siuu3t0013d000u000e8m\" width=\"1080\" height=\"512\" \/><\/p>\n<p>Related instructions: If ComfyUI is not the latest version, you need to update ComfyUI to the latest version first.<\/p>\n<p><strong>1. Plug-in installation<\/strong><\/p>\n<blockquote>\n<ul>\n<li>Plugin address: https:\/\/github.com\/city96\/ComfyUI-GGUF<\/li>\n<\/ul>\n<\/blockquote>\n<p>Search for ComfyUI-GGUF via the Plugin Manager.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18746\" title=\"5737626ej00siuu3s000ad000u0001wm\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/5737626ej00siuu3s000ad000u0001wm.jpg\" alt=\"5737626ej00siuu3s000ad000u0001wm\" width=\"1080\" height=\"68\" \/><\/p>\n<p>Simply click the install button to complete the installation.<\/p>\n<p>You can also use the git command to install it locally in the \\ComfyUI\\custom_nodes directory.<\/p>\n<blockquote>\n<ul>\n<li>git clone https:\/\/github.com\/city96\/ComfyUI-GGUF<\/li>\n<\/ul>\n<\/blockquote>\n<p>After the installation is complete, restart ComfyUI.<\/p>\n<p><strong>2. Download the flux1-dev GGUF model (you can also get the network disk address at the end of the article)<\/strong><\/p>\n<blockquote>\n<ul>\n<li>https:\/\/huggingface.co\/city96\/<a href=\"https:\/\/www.1ai.net\/en\/tag\/flux-1\" title=\"[SEE ARTICLES WITH [FLUX.1] LABELS]\" target=\"_blank\" >FLUX.1<\/a>-dev-gguf\/tree\/main<\/li>\n<\/ul>\n<\/blockquote>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18745\" title=\"a9486a5ej00siuu3t001nd000u000gfm\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/a9486a5ej00siuu3t001nd000u000gfm.jpg\" alt=\"a9486a5ej00siuu3t001nd000u000gfm\" width=\"1080\" height=\"591\" \/><\/p>\n<p><strong>Place the downloaded large model in the ComfyUI\\models\\unet directory.<\/strong><\/p>\n<ul>\n<li>For dev version, recommended: flux1-dev-Q8_0.gguf<\/li>\n<li>For channel version recommendation: flux1-dev-Q4_K_S.gguf<\/li>\n<\/ul>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18748\" title=\"ea996402j00siuu3s000ad000id002um\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/ea996402j00siuu3s000ad000id002um.jpg\" alt=\"ea996402j00siuu3s000ad000id002um\" width=\"661\" height=\"102\" \/><\/p>\n<p><strong>3. Download CLIP model<\/strong><\/p>\n<blockquote>\n<ul>\n<li>https:\/\/huggingface.co\/comfyanonymous\/flux_text_encoders\/tree\/main<\/li>\n<\/ul>\n<\/blockquote>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18749\" title=\"84984e97j00siuu3t000qd000u0008cm\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/84984e97j00siuu3t000qd000u0008cm.jpg\" alt=\"84984e97j00siuu3t000qd000u0008cm\" width=\"1080\" height=\"300\" \/><\/p>\n<p>Download t5xxl_fp8_e4m3fn.safetensors(4.89 GB) and clip_l.safetensors(246 MB) files and place them in \\ComfyUI\\models\\clip directory.<\/p>\n<p><strong>4. Download the VAE model<\/strong><\/p>\n<blockquote>\n<ul>\n<li>https:\/\/huggingface.co\/black-forest-labs\/<a href=\"https:\/\/www.1ai.net\/en\/tag\/flux\" title=\"_OTHER ORGANISER\" target=\"_blank\" >FLUX<\/a>.1-dev\/tree\/main<\/li>\n<\/ul>\n<\/blockquote>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18750\" title=\"32ea7c75j00siuu3t0012d000u000c2m\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/32ea7c75j00siuu3t0012d000u000c2m.jpg\" alt=\"32ea7c75j00siuu3t0012d000u000c2m\" width=\"1080\" height=\"434\" \/><\/p>\n<p>Download the ae.safetensors(335M) file and place it in the \\ComfyUI\\models\\vae directory.<\/p>\n<p><strong>5. Download the LORA model<\/strong><\/p>\n<blockquote>\n<ul>\n<li>https:\/\/huggingface.co\/XLabs-AI\/flux-lora-collection\/tree\/main<\/li>\n<\/ul>\n<\/blockquote>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18751\" title=\"5bafb413j00siuu3t001vd000u000cvm\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/5bafb413j00siuu3t001vd000u000cvm.jpg\" alt=\"5bafb413j00siuu3t001vd000u000cvm\" width=\"1080\" height=\"463\" \/><\/p>\n<p>Currently, Flux.1 has released 7 loars, including animation, art, Disney, MJV6, plush, realistic, and landscape. Download the required loar model files and place them in the ComfyUI\\models\\loras directory.<\/p>\n<p><strong>6. Use workflow<\/strong><\/p>\n<p>Click ComfyUI to start the script file, click \"load\" on the main interface and load the workflow file\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18752\" title=\"f54d9ecej00siuu3t0011d000u000chm\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/f54d9ecej00siuu3t0011d000u000chm.jpg\" alt=\"f54d9ecej00siuu3t0011d000u000chm\" width=\"1080\" height=\"449\" \/><\/p>\n<p>At this point, the ComfyUI workflow of Flux.1 dev GGUF can be applied.<\/p>\n<p><strong>Related Notes<\/strong>: The main difference between Flux.1 dev and Flux.1 dev GGUF workflows.<\/p>\n<p>Using the Load Diffusion Model node in the Flux.1 dev workflow<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18753\" title=\"28698580j00siuu3s0008d000ca004ym\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/28698580j00siuu3s0008d000ca004ym.jpg\" alt=\"28698580j00siuu3s0008d000ca004ym\" width=\"442\" height=\"178\" \/><\/p>\n<p>Flux.1 dev GGUF workflow using Unet Loader (GGUF) node<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-18754\" title=\"ea56064fj00siuu3t0005d000ae003am\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/08\/ea56064fj00siuu3t0005d000ae003am.jpg\" alt=\"ea56064fj00siuu3t0005d000ae003am\" width=\"374\" height=\"118\" \/><\/p>\n<p>Okay, that\u2019s all for today\u2019s sharing. I hope that what I shared today will be helpful to you.<\/p>\n<p>The model and workflow are placed in the network disk, and those who are interested can take them!<\/p>\n<p>https:\/\/pan.quark.cn\/s\/b3df771404e2<\/p>\n<\/div>\n<\/div>","protected":false},"excerpt":{"rendered":"<p>Previously shared with you some ways to experience Flux models. Flux [Basics]: share some websites where you can experience Flux.1 models online Flux [Basics]: local deployment and installation tutorial for ComfyUI Flux.1 workflow Today we are going to share how to deploy ComfyUI Flux.1 dev GGUF workflow locally. I. A First Look at GGUF GGUF is known as GPT-Generated Unified Format, a binary file format designed for large language models, proposed by Georgi Gerganov, the founder of llama.cpp, for efficiently storing and exchanging the pre-training results of large models. This format is optimized for<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[149,144],"tags":[1989,3853,3980,3878,1138],"collection":[],"class_list":{"0":"post-18742","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"hentry","6":"category-jiaocheng","7":"category-baike","8":"tag-comfyui","9":"tag-flux","10":"tag-flux-1","12":"tag-1138"},"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/18742","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=18742"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/18742\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=18742"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=18742"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=18742"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=18742"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}