comfyanonymous / comfyui_examples Goto Github PK

it is missing in the workflow and the manager can not find it. I found the json file https://huggingface.co/stabilityai/sdxl-turbo/commit/3a98214d01b1e1f5b0e71bec85ffb07534300fc7

but do not know how to install the node. update never mind

Allo!

I am beginning to work with ComfyUI moving from a1111 - I know there are so so many workflows published to civit and other sites- I am hoping to find a way to dive in and start working with ComfyUI without wasting much time with mediocre/redundant workflows and am hoping someone can help me by pointing be toward a resource to find some of the better developed comfy workflows

I would like to find workflows for SD 1.5 that allow me to work with img2img with masking, multiple controlnets, inpainting etc

Can anyone recommend any resources or specific workflows that are good for these tasks? OR can anyone recommend a resource/chat group/discord/repository that is good for finding the best comfyUI workflows? OR provide some tips how to sift through the ocean of workflows that are published online so that I am not wasting my time with mediocre implementations of various SD processes and am instead working with the best+most versatile existing ComfyUI implementations?

Any tips or advice would be infinitely appreciated! Thank you so much🙌🙌

I don't know because the other day updating, I stop working almost everything, I didn't work anymore the node created MP4, I kill my iPadapter, (from the workflow that I made to cheer up a video changing face and clothes,) I didn't walk the sdurb , the scheduler, quite bad everything is, I have been able to solve some things,
The Basic Sheduler served me to fix the Sigma, but what happened to the sdurboschedulate? =?
The iPadapter could fix it, I found in customscrips that appeared folders with direct access to who knows where,
I think some people who usually raise comfyui content, are sabotizing the things they had previously shared, for example some have gits with workflows,
Mysteriously take them away, (I don't know if they started walking or wanted to remove them)
Then other things go up, but they don't work, and derrepent, it doesn't go. is suspicious ..

Not sure whats making these images return blanks (not always but every now and then). Any suggestions?
I know its not the models i am using, most likely is the seed or noise - but i don't understand this enough.

The folder is missing the actual workflow that you could replicate/copy here. Probably the images placed there are not correct?

https://github.com/comfyanonymous/ComfyUI_examples/tree/master/sdxl

I get the following error while loading the template：

When loading the graph, the following node types were not found:
GLIGENTextBoxApply
GLIGENLoader
Nodes that have failed to load will show as red on the graph.

No issue!

I made some templates, but I wanted you to see it first before I posted it anywhere. I can change the name of the repo if you want. https://github.com/atlasunified/Templates-ComfyUI-

When starting Comfy the examples custom node does not import. The init.py file cannot be found.

Users/studiomaster/AI/ComfyUI/custom_nodes/ComfyUI_examples module for custom nodes: [Errno 2] No such file or directory: '/Users/studiomaster/AI/ComfyUI/custom_nodes/ComfyUI_examples/init.py'

Assuming i want to work with a large 16:9 images with many people in it, i think the area composition is what i want to use? is it compatible with controlnet.

I'm not familiar with a all possible kinds of loras but ones that I use didn't work until I added <lora:suzune-nvwls-v2-final:0.9> to prompt text, it's obvious to anyone who used a1111 before but ComfyUI example covers only adding LoraLoader and don't mention anything about prompt. I think it should be fixed.

Trying to use this example:
https://github.com/comfyanonymous/ComfyUI_examples/blob/master/3d/README.md

What node is missing and where to get it?

It's called Zero123_Conditioning
(Manager doesn't find it)

If I understand correctly there was a bug in the original 1.0 release.
They reverted the vae to 0.9, but they did not update the model it seems, so that the baked in vae in the model file is incorrect.
This can be solved by simply loading the vae file separately it seems...

https://www.reddit.com/r/StableDiffusion/comments/15ao8v7/sdxl_10_on_comfyui_default_workflow_weird_color/

RuntimeError: The expanded size of the tensor (1024) must match the existing size (768) at non-singleton dimension 0. Target sizes: [1024]. Tensor sizes: [768]

like change load models from \models\checkpoints to models\stable diffusion. Because I dont have that much disk space for copies of checkpoints