Instructions to use circlestone-labs/Anima with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusion Single File
How to use circlestone-labs/Anima with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Any clues for a new version of anima
#190
by KM-Kuro - opened
hey circlestone labs.
first of all, thanks for bring an amazing model which is 10 times better than SDXL models.
Is there any clues for a new version of anima with a more updated dataset, more characters and more styles that we can try and play around
They have a Anima Turbo version coming soon so I assume that would be the next version. Maybe the next year they release anima based on Nivida Cosmos 3 architecture. I don't know just my guess. They've done a super great job tho.
i hope so
Don't expect anything else on the horizon until:
- Anima 1.0 Turbo is finished. Should be twice as fast as Base but will not support negative prompts.
- Nvidia releases the previously announced Cosmos 3 Edge 4B (2B diffusion model with a 2B encoder). It's still in the oven and/or potentially delayed (though there was no specific release date given). Once out, it would quite literally be 4x faster and 4x easier for Circlestone to play around with than the current smallest Nano 16B.
- Just maybe... they make their Discord server public? π
I hope they change architecture because the current one sucks.