ComfyUI
ComfyUI copied to clipboard
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
Hi, first thanks a lot for your work! The visual form of the pipeline helped me a lot to better understand some basic concepts of SD! **My challenge:** I would...
Idk what im doing wrong. Im using protogen fp16 with fp16 yaml. Then tried img2img, but all i got is this. Prompt "Pretty girl" or something. The most coherent thing...
I am getting this error when using the example at: https://comfyanonymous.github.io/ComfyUI_examples/controlnet/mixing_controlnets.png The only changes I made are changing the controlnet model names to the names of my files, swapping to...
    Uses HSL to generate coherent colors easily. This is done procedurally so it won't need any update when a new category/output/input is...
I think I remember somewhere you were looking into supporting tensortRT models, is that still in the backlog somewhere? or would implementing support for tensortRT require too much rework of...
paperspace shows not url after running the collab.
I'd like to see intermediate steps so I know what I'm doing. Isn't it possible to allow VAE's or KSamplers to have the ability to show a preview of what...
Hello, As the title says. Loading a picture for the UI, ControlNET... etc, mean you have to move into folders each time. It's really really irritating. (Great UI gg)
I really think your actualy have the best UI, so if you want : 1-input text boxes (e.g. the prompt) have a problem: when you select the text and move...
Will Xformer be added in the future to solve the vrom usage problem?