:hugs: Diffusers provides pretrained diffusion models across multiple modalities, such as vision and audio, and serves as a modular toolbox for inference and training of diffusion models. PyPI: [https://pypi.org/project/diffusers/](https://pypi.org/project/diffusers/)