The statistical software package pySELFI implements the Simulator Expansion for Likelihood-Free Inference (SELFI) algorithm. SELFI is part of the family of simulation-based inference (SBI) methods, which replace the use of the likelihood function with a data-generating "black-box" simulator. pySELFI supports different black-box simulators, optimizes prior hyperparameters for primordial power spectrum inference, and check for model misspecification using the Mahalanobis distance. The code also scores compression and calculation of Fisher-Rao distances.