Pretrained diffusion models across multiple modalities using diffusers


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.

More precisely, 🤗 Diffusers offers:


  • State-of-the-art diffusion pipelines that can be run in inference with just a couple of lines of code (see src/diffusers/pipelines). Check this overview to see all supported pipelines and their corresponding official papers.
  • Various noise schedulers that can be used interchangeably for the prefered speed vs. quality trade-off in inference (see src/diffusers/schedulers).
  • Multiple types of models, such as UNet, can be used as building blocks in an end-to-end diffusion system (see src/diffusers/models).
  • Training examples to show how to train the most popular diffusion model tasks (see examplese.g. unconditional-image-generation).

Read the full article at: