Consider supporting further research on Patreon or Twitter
Checkout my other models at RetroDiffusion
Trained over SD v1.5
Used best with Dreamshaper v3.3 Baked VAE and v5
Use it at 0.85-0.8 of strength
Better results at 512x512, 512x768 and 768x512
Use simple prompts and no trigger word, see examples for prompt examples
You can control the color just by adding it to the prompt: picture of a corgi, red
At full strength, results are "burned"
It tend to add wings to everything, this is due to bad captioning, will be fixed on next releases
v1: Trained with raw dataset and raw captions, small dataset