This workflow simply
and generates an image.
Use this workflow only if you are sure the base checkpoint embeds a good quality VAE, otherwise check out another this workflow with VAE - https://openart.ai/workflows/openart/basic-sd15-workflow-with-vae/XlPv7Qz9U3wOIDn5vSxM.
", "created_at":{"_seconds":1671396844,"_nanoseconds":790000000}, "categories":[], "thumbnails":[{"width":960,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_OOJIcZFT_1702943704029_raw.jpg","height":960},{"width":2662,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_T1vSgAYq_1702943704333_raw.jpg","height":1530}], "nodes_count":{"total":9,"primitive":2,"custom":7}},{"id":"XlPv7Qz9U3wOIDn5vSxM","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1705490941,"_nanoseconds":409000000},"stats":{"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_comments":1,"num_likes":8,"num_downloads":1186,"num_views":4578},"nodes_index":["Note",".","ComfyUI","CLIPTextEncode","VAEDecode","SaveImage","EmptyLatentImage","KSampler","CheckpointLoaderSimple","VAELoader",","],"name":"Basic SD1.5 + VAE Workflow","description":"This workflow adds an external VAE on top of the basic text-to-image workflow ( https://openart.ai/workflows/openart/basic-sd15-workflow/lkOtNJ2UexVd6vK0kYhd). The VAE model is responsible of converting the latent image into the pixel space. The process is \"lossy\" and a good VAE model is essential to get good quality images.
This one workflow uses a default good VAE but there are more. Check out the description on Huggingface or CivitAI if the author suggests a specific VAE.
","created_at":{"_seconds":1671397200,"_nanoseconds":737000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/webp_3C6Rr-LO_1702933751971_raw.webp","height":1024},{"width":2620,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_i-o0SHeN_1702933984215_raw.jpg","height":1660}],"nodes_count":{"total":11,"primitive":3,"custom":8}},{"id":"7YARVB8mANBPGbzqpWfV","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671397911,"_nanoseconds":49000000},"stats":{"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_comments":1,"num_likes":12,"num_downloads":1369,"num_views":6916},"nodes_index":["Note",".","ComfyUI","CLIPTextEncode","VAEDecode","SaveImage","EmptyLatentImage","VAELoader","KSampler","CheckpointLoaderSimple","LoraLoader",","],"name":"Basic LoRA Workflow","description":"This workflow loads an additional LoRA on top of the base model.
** LORAs can be daisy-chained! You can have as many as you want **
","created_at":{"_seconds":1671397911,"_nanoseconds":900000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/webp_v5L3bGqX_1702941891063_raw.webp","height":1024}],"nodes_count":{"total":10,"primitive":1,"custom":9}},{"id":"qijtHYoTK3jG7grl0Mub","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_1024.jpg","username":"openart"},"updated_at":{"_seconds":1671398400,"_nanoseconds":796000000},"stats":{"num_comments":0,"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_likes":2,"num_downloads":665,"num_views":2314},"nodes_index":["Note",".","ComfyUI","VAEDecode","SaveImage","ImageScale","CLIPTextEncode","VAEEncode","RepeatLatentBatch","LoadImage","CheckpointLoaderSimple","KSampler",","],"name":"Basic SD 1.5 Img2img Workflow","description":"This is a basic img2img workflow. You can upload a reference image and a prompt to guide the image generation.
By default, it generates 4 images based on 1 reference image, but you can bypass or remove the Repeat Latent Batch node to generate just 1 image.
","created_at":{"_seconds":1671398400,"_nanoseconds":588000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_RYWWTOHy_1704183475455_raw.jpg","height":1024},{"width":3194,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_c3hZ3YG1_1704183518395_raw.jpg","height":1630}],"nodes_count":{"total":13,"primitive":3,"custom":10}},{"id":"YibsziaM9wW4Oze1iDwP","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671399000,"_nanoseconds":424000000},"stats":{"num_comments":0,"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_likes":1,"num_downloads":929,"num_views":3184},"nodes_index":["Note",".","ComfyUI","CLIPTextEncode","EmptyLatentImage","CheckpointLoaderSimple","KSampler","VAELoader","VAEDecode","PreviewImage","SaveImage","LatentFromBatch",","],"name":"Batch Generation Workflow","description":"This workflow can batch generate images instead of one at at time, by using the Latent From Batch node. It helps get more results at a time.
You can change the batch index to tweak the number of generations.
","created_at":{"_seconds":1671399000,"_nanoseconds":578000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_qmd3m0_k_1702943062775_raw.jpg","height":1024},{"width":3026,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_70ykMqSb_1702943063753_raw.jpg","height":1652}],"nodes_count":{"total":12,"primitive":1,"custom":11}},{"id":"S6PuWxBpXPB9PxSMhgUb","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_1024.jpg","username":"openart"},"updated_at":{"_seconds":1671399600,"_nanoseconds":439000000},"stats":{"num_comments":0,"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_likes":0,"num_bookmarks":0,"num_downloads":546,"num_views":2395},"nodes_index":["Note",".","ComfyUI","CLIPTextEncode","VAEDecode","SaveImage","EmptyLatentImage","VAELoader","KSampler","CheckpointLoaderSimple","CLIPSetLastLayer",","],"name":"Clip Skip Workflow","description":"This workflow allows you to skip some of the layers of the CLIP model when generating images. This can be useful for getting more creative results, as the CLIP model can sometimes be too specific in its descriptions.
","created_at":{"_seconds":1671399600,"_nanoseconds":951000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_8v6AdCia_1703246910524_raw.jpg","height":1024},{"width":3402,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_io5Y2QnI_1703246911529_raw.jpg","height":1678}],"nodes_count":{"total":10,"primitive":1,"custom":9}},{"id":"VSXSMBLJzsoy0Jb8qLIV","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671400200,"_nanoseconds":806000000},"stats":{"num_comments":0,"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_likes":4,"num_downloads":840,"num_views":2909},"nodes_index":["Note",".","ComfyUI","CLIPTextEncode","CheckpointLoaderSimple","EmptyLatentImage","KSampler","VAELoader","VAEDecode","PreviewImage","SaveImage",","],"name":"Basic Second-Pass Workflow","description":"This workflow applies a second pass of KSampler on top of this basic SD 1.5 workflow with an external VAE ( https://openart.ai/workflows/openart/basic-sd15-workflow-with-vae/XlPv7Qz9U3wOIDn5vSxM). Even at the same resolution (ie: without upscaling), the second pass can increase definition and fix small mistakes.
Increasing the DENOISE adds more details but also changes the original reference more (the first pass).
This inpainting workflows allow you to edit a specific part in the image. Right click the image, select the Mask Editor and mask the area that you want to change. The grow mask option is important and needs to be calibrated based on the subject.
In this example we're applying a second pass with low denoise to increase the details and merge everything together.
","created_at":{"_seconds":1671400800,"_nanoseconds":920000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_sDsn1iWt_1703246015468_raw.jpg","height":1024},{"width":3282,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_pNgw9yvC_1703246016572_raw.jpg","height":1736}],"nodes_count":{"total":14,"primitive":4,"custom":10}},{"id":"4UymtsV7TcfpZPHPgIpd","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_1024.jpg","username":"openart"},"updated_at":{"_seconds":1671401400,"_nanoseconds":622000000},"stats":{"num_comments":0,"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_likes":3,"num_downloads":569,"num_views":4176},"nodes_index":["Reroute","Note",".","ComfyUI","CLIPTextEncode","VAELoader","KSampler","VAEDecode","ImagePadForOutpaint","CheckpointLoaderSimple","PreviewImage","VAEEncodeForInpaint","EmptyLatentImage",","],"name":"Basic Outpainting Workflow","description":"In this workflow, the first half of the workflow just generates an image that will be outpainted later. You can replace the first with an image import node.
In the second half othe workflow, all you need to do for outpainting is to pad the image with the \"Pad Image for Outpainting\" node in the direction you wish to add. Note that it's still technically an \"inpainting\" process, so you need to use an inpainting checkpoint and the \"VAE Encode (for inpainting)\" node.
","created_at":{"_seconds":1671401400,"_nanoseconds":823000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_2xx-i5SI_1704184168022_raw.jpg","height":1024},{"width":3258,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_xDFw9YQO_1704184171763_raw.jpg","height":1394}],"nodes_count":{"total":19,"primitive":5,"custom":14}},{"id":"aO8mb2DFYJlyr7agH7p9","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_1024.jpg","username":"openart"},"updated_at":{"_seconds":1671402000,"_nanoseconds":214000000},"stats":{"num_shares":0,"rating":5,"num_runs":0,"num_bookmarks":0,"num_reviews":1,"num_comments":1,"num_likes":5,"num_downloads":447,"num_views":1440},"nodes_index":["Reroute","Note",".","ComfyUI","CLIPTextEncode","VAEDecode","EmptyLatentImage","SolidMask","FeatherMask","PreviewImage","KSampler","MaskComposite","MaskToImage","SetLatentNoiseMask","ImagePadForOutpaint","SaveImage","VAELoader","VAEEncodeForInpaint","CheckpointLoaderSimple",","],"name":"Outpainting With Seam Fix","description":"When outpainting it happens to get a noticeable seam line between the two pieces. This workflow applies a low denoise second pass over the outpainted image to fix any glitch.
","created_at":{"_seconds":1671402000,"_nanoseconds":523000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_VraAqcdy_1704185036207_raw.jpg","height":1024},{"width":3316,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_5gWQrT5v_1704185041158_raw.jpg","height":1630}],"nodes_count":{"total":30,"primitive":7,"custom":23}},{"id":"tOYu3yw8kSXxA1gB96Db","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671404400,"_nanoseconds":77000000},"stats":{"num_comments":0,"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_likes":4,"num_downloads":559,"num_views":5962},"nodes_index":["Note",".","ComfyUI","EmptyLatentImage","VAEDecode","CLIPTextEncode","KSampler","SaveImage","CheckpointLoaderSimple",","],"name":"Ultra Basic SDXL Workflow (Not Recommended)","description":"This is essentially the same as the basic workflow for SD 1.5 ( https://openart.ai/workflows/openart/basic-sd15-workflow/lkOtNJ2UexVd6vK0kYhd ), but it's not optimized for SDXL. We include it here for the demo purpose.
We highly recommend you to use another basic SDXL workflows - https://openart.ai/workflows/openart/basic-sdxl-workflow/P8VEtDSQGYf4pOugtnvO which includes ClipTextEncodeSDXL for better image quality results.
","created_at":{"_seconds":1671404400,"_nanoseconds":591000000},"categories":[],"thumbnails":[{"width":768,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_4BaFHbOX_1702943833144_raw.jpg","height":768},{"width":2744,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_DJNqO6SF_1702943833773_raw.jpg","height":1606}],"nodes_count":{"total":9,"primitive":2,"custom":7}},{"id":"P8VEtDSQGYf4pOugtnvO","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671405000,"_nanoseconds":612000000},"stats":{"num_comments":0,"num_shares":0,"num_runs":0,"num_bookmarks":0,"num_reviews":1,"rating":5,"num_likes":4,"num_downloads":1699,"num_views":7115},"nodes_index":["Note","PrimitiveNode",".","ComfyUI","CheckpointLoaderSimple","CLIPTextEncodeSDXL","KSampler","EmptyLatentImage","VAEDecode","SaveImage",","],"name":"Basic SDXL Workflow","description":"This basic workflow runs the base SDXL model with some optimization for SDXL. This can be useful for systems with limited resources as the refiner takes another 6GB or ram.
If your model is based on SD 1.5, use this basic workflow instead - https://openart.ai/workflows/openart/basic-sd15-workflow/lkOtNJ2UexVd6vK0kYhd.
This workflow adds a refiner model on topic of the basic SDXL workflow ( https://openart.ai/workflows/openart/basic-sdxl-workflow/P8VEtDSQGYf4pOugtnvO ).
The core of the composition is created by the base SDXL model and the refiner takes care of the minutiae. The refiner helps improve the quality of the generated image.
","created_at":{"_seconds":1671405600,"_nanoseconds":923000000},"categories":["SDXL"],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_zQVjBhMQ_1704182787567_raw.jpg","height":1024},{"width":2662,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_0IHo0xiZ_1704182789124_raw.jpg","height":1402}],"nodes_count":{"total":22,"primitive":10,"custom":12}},{"id":"yrUf3Yj2kqSHVbZsS9vy","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_1024.jpg","username":"openart"},"updated_at":{"_seconds":1671406200,"_nanoseconds":206000000},"stats":{"num_comments":0,"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_likes":9,"num_views":6538,"num_downloads":1574},"nodes_index":["PrimitiveNode","Note",".","ComfyUI","VAEDecode","KSampler","CLIPTextEncodeSDXL","ImageScale","LoadImage","CheckpointLoaderSimple","VAELoader","VAEEncode","SaveImage",","],"name":"Basic SDXL Img2Img Workflow","description":"This is a basic img2img workflow on top of our basic SDXL workflow ( https://openart.ai/workflows/openart/basic-sdxl-workflow/P8VEtDSQGYf4pOugtnvO ). Instead of starting with a random latent image, the workflow will start with a user-uploaded image.
","created_at":{"_seconds":1671406200,"_nanoseconds":253000000},"categories":["SDXL"],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_skZX6-ue_1703244597599_raw.jpg","height":1024},{"width":3228,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_GbcOxaL6_1703244598402_raw.jpg","height":1710}],"nodes_count":{"total":15,"primitive":5,"custom":10}},{"id":"REGMkdFKDcRg4RkpMzvG","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671408000,"_nanoseconds":294000000},"stats":{"num_shares":0,"rating":5,"num_runs":0,"num_bookmarks":0,"num_comments":1,"num_reviews":1,"num_likes":13,"num_downloads":1494,"num_views":14772},"nodes_index":["Note",".","ComfyUI","CLIPTextEncode","EmptyLatentImage","KSampler","VAELoader","VAEEncode","VAEDecode","SaveImage","ImageScaleBy","PreviewImage","CheckpointLoaderSimple",","],"name":"Pixel Upscale Workflow","description":"This is the simplest and probably less expensive method to upscale an image on the pixel level. A lanczos algorithm is used in the pixel space to upscale the image and then a second pass applied to it.
It needs a relatively high denoise which means the final image will be a little different from the source.
","created_at":{"_seconds":1671408000,"_nanoseconds":124000000},"categories":[],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/webp_dq9KHLx1_1702944090473_raw.webp","height":1024},{"width":3152,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_E0ZP2RF-_1702944790937_raw.jpg","height":1644}],"nodes_count":{"total":14,"primitive":1,"custom":13}},{"id":"V6horzh1YQN8Pz5rJXzP","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671408600,"_nanoseconds":974000000},"stats":{"num_shares":0,"rating":5,"num_reviews":0,"num_runs":0,"num_bookmarks":0,"num_likes":1,"num_comments":1,"num_downloads":1104,"num_views":5674},"nodes_index":["Note",".","ComfyUI","CLIPTextEncode","EmptyLatentImage","CheckpointLoaderSimple","KSampler","VAEDecode","SaveImage","VAELoader","PreviewImage","ImageUpscaleWithModel","UpscaleModelLoader","ImageScaleBy","VAEEncode",","],"name":"Model-based Pixel Upscale Workflow","description":"This workflows loads an existing upscaling model that takes care of the upscaling phase. It's more effective but more expensive than our basic image upscaling workflow ( https://openart.ai/workflows/openart/basic-image-upscale-workflow/REGMkdFKDcRg4RkpMzvG ).
The second pass is performed to add back some sharpness to the image but can be done at a very low denoise.
The latent can be upscaled directly but the loss in quality is important. That means that the second pass after the upscale has to be done at a very high denoise (0.55+).
This option is offered as it's very inexpensive computationally and it can be useful if you don't need the final image to be 100% of the initial generation.
If the final image can be rather different from the first pass, you should use latent upscale as it's faster and very light computationally.
","created_at":{"_seconds":1671409200,"_nanoseconds":933000000},"categories":["Upscale"],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_yiC2Eo7L_1704182268845_raw.jpg","height":1024},{"width":2958,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_GN_Yjfa5_1704182271315_raw.jpg","height":1608}],"nodes_count":{"total":14,"primitive":2,"custom":12}},{"id":"8H0mGG5dOFQCuslp2qQv","creator":{"uid":"4pYEgRw3JXVFu5elJKeR","name":"OpenArt","bio":"","username":"openart","avatar":"https://cdn.openart.ai/uploads/image_vmS7sTuQ_1703026397814_256.jpg"},"updated_at":{"_seconds":1671415200,"_nanoseconds":788000000},"stats":{"num_shares":0,"num_runs":0,"num_bookmarks":0,"rating":3,"num_reviews":1,"num_comments":1,"num_likes":17,"num_downloads":1864,"num_views":7213},"nodes_index":["Note",".","ComfyUI","EmptyLatentImage","CLIPTextEncode","VAEDecode","VAELoader","SaveImage","LoadImage","KSampler","CLIPVisionLoader","CheckpointLoaderSimple",",","ComfyUI_IPAdapter_plus","IPAdapterApply","IPAdapterModelLoader",","],"name":"Basic IPAdapter","description":"This workflows is a very simple workflow to use IPAdapter IP-Adapter is an effective and lightweight adapter to achieve image prompt capability for stable diffusion models.
The IPAdapter model has to match the CLIP vision encoder and of course the main checkpoint. All SD15 models and all models ending with \"vit-h\" use the SD15 CLIP vision. One of the SDXL models and all models ending with \"vit-g\" use the SDXL CLIP vision.
PLUS models use more tokens and are stronger. LIGHT models have a very light impact. FACE and FULL-FACE are only to describe faces (they are not face swap!)
","created_at":{"_seconds":1671415200,"_nanoseconds":93000000},"categories":["IP-Adapter"],"thumbnails":[{"width":1024,"url":"https://cdn.openart.ai/workflow_thumbnails/4pYEgRw3JXVFu5elJKeR/image_KdBztP1T_1702945457521_raw.jpg","height":1024}],"nodes_count":{"total":13,"primitive":1,"custom":12}}] export const modules = [ // Comfy Academy ...COURSES.map(course => ({ title: course.title, description: course.description, type: 'lesson', videoId:course.video_embed_link.substring(course.video_embed_link.lastIndexOf('?v=')+1), categories:[categories[0]], thumbnail:course.thumbnail_url, })), // templates ...TEMPLATES.map(template => ({ id: template.id, title: template.name, description: template.description, type: 'template', thumbnail:template.thumbnails[0].url, categories:[categories[1]], })) ] fetch("https://openart.ai/api/public/templates/list").then(res=>res.json()).then(res=>{ res.map(template => { const t = ({ id: template.id, title: template.name, description: template.description, type: 'template', thumbnail: template.thumbnails[0].url, categories: [categories[1]], }) if(!modules.find(m=>m.id===t.id)){ modules.push(t) } } )})