I've also tested this with some very complex workflows. Including a 15 minute AnimateDiff + IPAdapter + OpenPose video to video workflow. Thank you to @peterom
, @banodoco and @thibaudz for their examples.
@fofrai
-
Complex AnimateDiff IPAdapter OpenPose Video Workflow Testing
By
–
-
ComfyUI and ComfyDeploy: Open Source Production AI Deployment
By
–
I'd also like to point out that this is just one of the open source ways to use ComfyUI in production. @BennyKokMusic is doing great work on ComfyDeploy:
-
ComfyUI Workflow Model Available on Replicate Platform
By
–
Model: https://
replicate.com/fofr/any-comfy
ui-workflow
… Repo (request weights and custom nodes here): https://
github.com/fofr/cog-comfy
ui
… Full list of supported weights: https://
github.com/fofr/cog-comfy
ui/blob/main/supported_weights.md
… -
ComfyUI API Examples for Image and Video Generation
By
–
A variety of API examples are included in the repo: https://
github.com/fofr/cog-comfy
ui/tree/main/examples
… img2img with an input: https://
replicate.com/p/qp5kxkdb5vsx
vvrb7zkcsjspue
… PhotoMaker: https://
replicate.com/p/fmquundbilav
gmvyswtzynbtam
… 37 different controlnet preprocessors: https://
replicate.com/p/6g3pol3bawfx
3mjpleqd76h5hi
… AnimateDiff: https://
replicate.com/p/mz67hxdbjfpi
qk32kdnglbvdmi
… -
ComfyUI API Model on Replicate Supports Advanced AI Features
By
–
I've been working on a new model on Replicate that lets you run any ComfyUI workflow with an API.
— fofr (@fofrAI) 22 janvier 2024
It supports all the popular controlnets, base weights, preprocessors, photomaker, animatediff, LCM, upscalers, IPAdapters.
Details in 🧵 pic.twitter.com/VhHKEqEbCyI've been working on a new model on Replicate that lets you run any ComfyUI workflow with an API. It supports all the popular controlnets, base weights, preprocessors, photomaker, animatediff, LCM, upscalers, IPAdapters. Details in
-
UltimateSDupscaler Added to Comfy UI Integration
By
–
I've added all of these: https://
github.com/fofr/cog-comfy
ui/commit/56532c94fb5f0ac2587ba4822b4e55af93a15347
… UltimateSDupscaler should have been a given -
Weight-free AI model startup performance issue
By
–
Not sure what's happened there. It's a weight-free model so start up is fast. If you DM me the details I'll raise with team.
-
Midjourney v3 Output Discovered in v6 Release
By
–
I found a Midjourney v3 output in v6, instant favourite.
-
Native 2048x Fast Model Development Without Upscaling
By
–
I’m hoping for a native 2048x fast model, not an upscaler.
-
First Look at Stable Diffusion 3 Model Release
By
–
Is this our first look at Stable Diffusion 3? Emad hinted at this in July last year. My guess is that this is a new model not an SDXL iteration. Also quoting again because I like the new cat picture better.