Instructions to use xinsir/controlnet-tile-sdxl-1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use xinsir/controlnet-tile-sdxl-1.0 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("xinsir/controlnet-tile-sdxl-1.0", 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 Settings
- Draw Things
- DiffusionBee
Do you have the guide or tutorial for training this controlnet?
#4
by MorPhLingXD - opened
Thanks for the great work!
I am a beginning researcher and am interested in training a ControlNet tile model by myself. It is interesting to see a tile model can have different behaviors (e.g. deblur/vary the detail) so I believe there are some training schemes behind it.
In addition, I saw your union ControlNet model that combines the tile condition with other conditions. Would you release the details of training settings of it in your paper?
Bumping this for how to train a tile model