Color gelbooru images

The model is made up of components from Mage Flow, it colors the input image based on the color tags.

There are 1000 color-related matrices that can be used as conditions.

Keywords

  • pixel space model
  • U-shaped architecture
  • fully convolutional model
  • cross-attention between encoder and decoder blocks

Source data

  • 250k augmented, colored images
  • realistic render (similar to academic art movement)
  • dohna dohna
  • noripro
  • stellar sora
  • vspo!

References

  • 2204.00631
  • 2508.02324
  • 2511.13720
  • 2603.15031
  • 2607.19064
Downloads last month

-

Downloads are not tracked for this model. How to track
Safetensors
Model size
46.5M params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support