Instructions to use kandinsky-community/kandinsky-2-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use kandinsky-community/kandinsky-2-1 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", 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
Has anyone gotten this running on mps?
#7
by iandoesallthethings - opened
I've gotten all the way through downloading the models and it kicks back this error:
error: input types 'tensor<1x154x1xf16>' and 'tensor<1xf32>' are not broadcast compatible
But I'm pretty new to diffusers, so I'm guessing there's an incompatibility between the device and the model?