Download
2 variants available
fp16 SafeTensor
SDArt_v1.safetensors
Half precision, best balance (pruned) • 2.4 GB
Verified: 3 years ago
SafeTensor
Other Formats
This checkpoint includes a config file, download and place it along side the checkpoint.
2750 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
(27)
Apr 18, 2023
SD 2.1 768
First version, base 2.1. 14 different contests worth of datasets, 619 pictures.
Datasets included :
Themes trained :
Type of art (3D, abstract, painting, rendering, anime, ...)
Composition (asymmetrical, portrait, fractal, ...)
Colors (vivid colors, dark, colorful, rainbowshift, ...)
Feelings (cute, strange, calm, chaos)
Subjects (Alice, cats, Miku, mermaids, cows, tentacles, monster, creature, ...)
Backgrounds (landscapes, sky, night, space, trees, field, underwater, cyberpunk city, ...)
Show more

1590 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1.8K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
License:
CreativeML Open RAIL-MThis is a model trained on all the previous PoW datasets at once, using complete captions (at last) for better usability. It is related to the different contests that happen weekly on Stable Diffusion discord.
The current version, v2, has two versions itself, one trained from SD1.5, and one from SD2.1 768. 16 contests worths of pictures, 742 total dataset
Currently includes :
"Community call" dataset -> Examples with prompts
Datasets available on the individual models.
The v1 was only trained on 2.1 as base. 16 different contests worth of datasets, 619 pictures. It misses Community call, Unmythical Creatures, and Encapsulated.
Complete dataset and diffusers available on Hugging face, for the v2 base 1.5 and the v2 base 2.1 versions.

