A nude/NSFW capable model that focuses on images with a female primary subject with an amateur artistic aesthetic.
Recommended Settings
Sampler is dpmpp_2m (DPM++ 2M) and the scheduler is beta
No VAE or CLIP is baked in. Use separate sources for those.
Each sample/preview image contains the used workflow. Here's a quick article with simpler more beginner-friendly workflow. This is a recommended starting point.
Description
FAQ
Comments (7)
is this a checkpoint or diffusion model? (how can you even tell, I prob have a few in wrong place lol)
Yeah this is really annoying and I don't know of a way to tell!
This is a checkpoint! So it goes in checkpoints folder.
You have the best models but you didn't do GGUF Q4 for the latest updates, so I'm trying to figure out how to do them myself and the quantizer is simply crashing.
Yeat it's just too much work to quantize everything to everything. I rather spend my time working on more models than that usually.
Hi!
I'm trying to use your awesome amafluxAmateurRealistic_22FP16 model in ComfyUI for a project, but I keep getting tensor shape mismatch errors during the KSampler or VAE Decode stages. My other Flux2 workflows run perfectly, but this specific merge gives me a RuntimeError: The size of tensor a (16) must match the size of tensor b (128) error when decoding. On my system (RTX 5070 Ti, 64GB RAM), I am using separate t5xxl_fp16, clip_l, and flux2-vae sources along with your recommended dpmpp_2m/beta settings. Could you suggest how I should structure the node network to prevent these matrix dimension conflicts between the text encoders and the model? Alternatively, if you could share a screenshot or JSON of your own working ComfyUI workflow template for this model, it would be a huge help!
Thanx!
It's not a Flux 2 workflow. It's a Flux1 model. This is likely the problem :)
Try to use an included workflow, or a Flux 1 workflow.
@6tZ oh my God! what a shame to me! Sorry bud but thanks.