Instructions to use Yntec/VintedoisRemix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Yntec/VintedoisRemix with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Yntec/VintedoisRemix", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Vintedois Remix
I mixed 22h's Vintedois model with 22 different models in an attempt to improve it, and I think I did it with the help of realisticElves's animeTEN model! The idea was to create a model called animeTWENTYTWO with these models, but it kept improving as I kept in Vintedois's blocks, until just 3 from animeTEN did the trick! Work in progress.
Vintedois Remix Alpha
While Vintedois Remix maximized the amount of detail, Alpha keeps the detail at the level of Vintedois thanks to the CosineB merging technique. I figure that if you want to use these new models for merging you'd want to use this one instead (which may be better than Remix, anyway, when in doubt I always choose the most detailed model.)
Recipe
- SuperMerger Weight Sum (normal) Use MBW 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,1,0,0,0,0,0
Model A:
Vintedois
Model B:
animeTEN
Output:
VintedoisRemix
- SuperMerger Weight Sum (cosineB) Use MBW 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0.5,0,1,0,0,0,0,0
Model A:
Vintedois
Model B:
animeTEN
Output:
VintedoisRemixAlpha
- Downloads last month
- 7