Trained on images I like.
Seems to give me results i like at 0.7 - 1 (depending on prompt)
what_the_yell.safetensors
Same model published on other platforms. May have additional downloads or version variants.