scvi-tools (single-cell variational inference tools) [Gayoso21] is a package for end-to-end analysis of single-cell omics data primarily developed and maintained by the Yosef Lab at UC Berkeley. scvi-tools has two components

  • Interface for easy use of a range of probabilistic models for single-cell omics (e.g., scVI, scANVI, totalVI).

  • Tools to build new probabilistic models, which are powered by PyTorch, PyTorch Lightning, and Pyro.

If you find a model useful for your research, please consider citing the corresponding publication, which can be found in the corresponding model documentation.



New to scvi-tools? Check out the installation guide.


User guide

The user guide provides distilled mathematical descriptions of the models implemented in scvi-tools and connects the math with the code.


API reference

The API reference contains a detailed description of the scvi-tools API.



The tutorials walk you through real-world applications of scvi-tools models. Developer tutorials help you build new probabilistic models.