Hugging Face's logo

AbstractPhil
/
sd15-flow-lune

Text-to-Image
Diffusers
TensorBoard
Safetensors
stable-diffusion
diffusion
distillation
flow-matching
geometric-deep-learning
research
Model card Files
xet
 Metrics Community
AbstractPhil commited on
Commit
528181a
·
verified ·
1 Parent(s): 618c9e2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -1
README.md CHANGED
@@ -21,9 +21,12 @@ pipeline_tag: text-to-image
21
  Currently retraining the scale, but it was trained with many raw unscaled latents and it makes the default output hazy.
22
  ![image](https://cdn-uploads.huggingface.co/production/uploads/630cf55b15433862cfc9556f/6GFXrQy6vm8h2mdkK5mvD.png)
23
  Use this to correctly orient the output to the correct VAE scale.
24
- ![image](https://cdn-uploads.huggingface.co/production/uploads/630cf55b15433862cfc9556f/Q_bHlE8yv2bFLtQsixG_8.png)
25
 
 
 
 
26
 
 
27
  ![image](https://cdn-uploads.huggingface.co/production/uploads/630cf55b15433862cfc9556f/zXNIFANpK7Yqmm4oPUuUR.png)
28
 
29
 
 
21
  Currently retraining the scale, but it was trained with many raw unscaled latents and it makes the default output hazy.
22
  ![image](https://cdn-uploads.huggingface.co/production/uploads/630cf55b15433862cfc9556f/6GFXrQy6vm8h2mdkK5mvD.png)
23
  Use this to correctly orient the output to the correct VAE scale.
 
24
 
25
+ ## Shift 2 is the training target
26
+ ![image](https://cdn-uploads.huggingface.co/production/uploads/630cf55b15433862cfc9556f/3aUl0td4RiDL9yjMw87KT.png)
27
+ Higher or lower may yield different results.
28
 
29
+ ## use this
30
  ![image](https://cdn-uploads.huggingface.co/production/uploads/630cf55b15433862cfc9556f/zXNIFANpK7Yqmm4oPUuUR.png)
31
 
32