I haven’t touched this in nearly a year and I don’t plan on coming back to it at this point. Feel free to make your own fork if you want to add additional nodes or fixes.
ComfyUI-Cloud-APIs
I wanted to use some larger models in my local ComfyUI workflows but I’m too gpu poor. Currently model support is limited but I’ll add more when I find the time.
Supported Models:
Flux t2i (fal, replicate, runware)
Flux i2i (fal)
Flux w/loras and i2i (fal)
Auraflow t2i (fal)
SoteDiffusion t2i (fal)
StableCascade t2i (fal)
LLaVA 1.5 13B i2t (fal)
LLaVA 1.6 34B i2t (fal)
SDv1 t2i w/loras (runware)
SDXL t2i w/loras (runware)
Quick Setup
Install extension and dependencies (preferably with comfy manager)
Place your api key/token in a text file in ComfyUI-Cloud-APIs/keys
Step by step setup (Fal)
Install the extension with the “Install via Git URL” option in comfyUI manager
Go to https://fal.ai/dashboard/keys and click “Add key”
Name the key then copy it into a text file in the ComfyUI-Cloud-APIs/keys folder. (there is a placeholder nokey.txt file which you can delete)
Consult the models page to get an idea of how much each generation will cost
Go to https://fal.ai/dashboard/billing and top up your account. For your financial well-being I recommend against automated topups, but I can’t stop you.
Step by step setup (Replicate)
Install the extension with the “Install via Git URL” option in comfyUI manager
Create an account at https://replicate.com/
Go to https://replicate.com/account/api-tokens and copy your token (or create a new one)
Copy the token into a text file in the ComfyUI-Cloud-APIs/keys folder. (there is a placeholder nokey.txt file which you can delete)
Consult https://replicate.com/explore to get an idea of how much each generation will cost
Go to https://replicate.com/account/billing to setup billing when you run out of free usage.