Downloads · 30 days
21
6% of all-time downloads
sdxl-vae-midjourneyv6-ft-wrong
lavinal712/sdxl-vae-midjourneyv6-ft-wrong
sdxl-vae-midjourneyv6-ft-wrong is a image-to-image model from lavinal712. Use it when you need one image transformed into another. It is set up for diffusers. The card lists the license as mit.
All-time downloads
331
Public
Parameters
83.7M
338 MB on disk
Likes
4
Public
Repo makeup
Click a slice to open those files.
.safetensors335 MB · 99%
At a glance
- Task
- Image-to-Image
- Library
- diffusers
- License
- mit
- Access
- Public
- Created
- Mar 23, 2025
- Updated
- Mar 24, 2025
- SHA
- 05f31201
Base models
Datasets
Model card
From the Hugging Face model README
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 |
- Task
- Image-to-Image
- Library
- diffusers
- License
- mit
- Languages
- en
- Created
- Mar 23, 2025
- Updated
- Mar 24, 2025