Instructions to use IndexTeam/Index-anisora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use IndexTeam/Index-anisora with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("IndexTeam/Index-anisora", 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
Unofficial AniSora V3.2 endpoint on Replicate
Hi AniSora team — thanks for releasing AniSora. We’ve been experimenting with character turntables and image animation, and built an unofficial Replicate endpoint to make V3.2 easier to try.
Endpoint: https://replicate.com/sprited/anisora
Packaging code: https://github.com/sprited-ai/anisora-on-replicate
It uses Kijai’s FP8-scaled high- and low-noise weights through ComfyUI, with attribution to your project. We disclose the quantization since results may differ from the original full-precision weights.
Wanted to share it with you directly. If you have feedback on the implementation, naming, or attribution, we’d be happy to adjust it. Thanks again for releasing AniSora!
If you find it useful, would you consider adding a link to this unofficial endpoint in your GitHub README or Hugging Face model card? It could help more people try the model without setting up a GPU environment. More regular usage may also reduce cold-start waits on the hosted endpoint.