import torch
from diffusers import DiffusionPipeline
from diffusers.utils import load_image
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("lavinal712/sdxl-vae-midjourneyv6-ft-wrong", dtype=torch.bfloat16, device_map="cuda")
prompt = "Turn this cat into a dog"
input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat.png")
image = pipe(image=input_image, prompt=prompt).images[0]Finetune repo: lavinal712/AutoencoderKL
Dataset: CortexLM/midjourney-v6
Eval Dataset: playgroundai/MJHQ-30K
Fine-tuning modules: decoder and quant_conv
Input:
Reconstruction:
| metrics on ImageNet | rFID | PSNR | SSIM | LPIPS | gFID (MJHQ-30K) |
|---|---|---|---|---|---|
| sdxl-vae | 0.665 | 27.376 | 0.794 | 0.122 | 6.558 |
| finetuned (ours) | 1.358 | 27.590 | 0.802 | 0.121 | 6.643 |
- Downloads last month
- 9
Model tree for lavinal712/sdxl-vae-midjourneyv6-ft-wrong
Base model
stabilityai/sdxl-vae