How do you get this to work?
I put it in the checkpoint folder but it didnt work, is there a different installation method?
You have to put it in the "unet" folder instead
You have to put it in the "unet" folder instead
where is that folder?
Are you using ComfyUI or just diffusers library?
Did you found a solution? Same problem here, tried to loadcheckpoint and gave me this error:
Error occurred when executing CheckpointLoaderSimple:
ERROR: Could not detect model type of: I:\Comfy UI\ComfyUI\models\checkpoints\flux_dev.safetensors
It needs to be placed in the unet folder . It can be found in the models folder. It wont work as a normal checkpoint.
Did you found a solution? Same problem here, tried to loadcheckpoint and gave me this error:
Error occurred when executing CheckpointLoaderSimple:
ERROR: Could not detect model type of: I:\Comfy UI\ComfyUI\models\checkpoints\flux_dev.safetensors
move it to I:\Comfy UI\ComfyUI\models\unet
There are a few 16gb "all in one" Flux checkpoints on CivitAI that have the text decoder and vae files merged into them. They allegedly use less VRAM and produce images faster. The only instruction provided is to put them in the "checkpoint" folder (not "unet"). But I can't find a Flux workflow to make use of them in ComfyUI since all the Flux workflows pull the separate files form the unet, vae and clip folders.