5/24/2026 - Updated v3.0 file to fix a PyTorch size mismatch error.
Grounded generates realistic images of realistic women. Not another merge of merges, Grounded is made from trained checkpoints, particularly ones that specialize in amateur photography, and, therefore, amateur women. It excels in believable faces, realistic breasts, and diverse body types.
All samples were generated without embeddings, loras, or upscaling.
Recommendations:
Clip skip: 1-2
CFG: 6-8 (4-6 for v3.0)
20-40 steps
512-768
DPM++ 2M Karras, DPM++ SDE Karras, UniPC
Avoid negative embeddings
Description
Back back, back back, back again!
FAQ
Comments (12)
this is really impressive. it really does show the small imperfections that people have, like for example an eye lid that is slightly lower than its counterpart
for some strange reason I can't load it
same, getting this exception-error at start in Comfyui, 'NoneType' object has no attribute 'supported_inference_dtypes'
Thanks for letting me know. I hate to hear it. Unfortunately, I do and always have done everything in the Draw Things iPhone app. I’ve downloaded and reinstalled this safetensors file, and it’s working fine for me. I don’t have a PC or any other platforms to test on. But I will try to see if I can find something out. Thank you.
I'm getting this error with the latest 1.7 version: "size mismatch for model.diffusion_model.output_blocks". Previous version 1.6 works well
EDIT: 1.7 removed for the time being.
_____
I’ve received a few reports of errors installing version 1.7. I’m not sure how widespread this issue is, nor do I have familiarity with ComfyUI or Forge. Best I can figure is to try uploading the full unpruned version, which I will do this evening.
Sorry, i tried the latest version - 38 min ago uploaded - around 4GB - it also fires the error(s).
I confirm - an error occurs (AUTOMATIC1111/stable-diffusion-webui ):
RuntimeError: Error(s) in loading state_dict for LatentDiffusion:
size mismatch for model.diffusion_model.input_blocks.1.1.proj_in.weight: copying a param with shape torch.Size([320, 320]) from checkpoint, the shape in current model is torch.Size([320, 320, 1, 1]).........
same result, getting "AttributeError: 'NoneType' object has no attribute 'supported_inference_dtypes'". Using comfyui latest version I even upgraded to python's inference 0.11.2 but doesnt seem related to the error
Thanks for letting me know, y’all. I’ll try to figure out what these error mean, this weekend. It’s been a very busy week for me.