BayesFlow

Spec sheet

What it does

BayesFlow is a Python library for simulation-based amortized Bayesian inference with neural networks.Read 11 Oct 2026 · bayesflow.org

Purpose

It trains neural networks on simulated datasets to produce Bayesian quantities such as posterior samples, likelihoods, likelihood ratios, point estimates, and model probabilities.Read 11 Oct 2026 · bayesflow.org

Product

BayesFlow is a Python library for simulation-based amortized Bayesian inference with neural networks.Read 11 Oct 2026 · bayesflow.org

Deployment

desktopRead 23 Sep 2026 · bayesflow.org

Free plan

YesRead 23 Sep 2026 · bayesflow.org

Integrations

Optional wrappers integrate Mamba summary networks through mamba-ssm and allow BayesFlow approximators to be used as likelihood terms in PyMC models.Read 11 Oct 2026 · bayesflow.org

Support

The project directs users to its Discourse forums for questions and GitHub Issues for bug reports and feature requests.Read 11 Oct 2026 · bayesflow.org

License

BayesFlow is released under the MIT License and its source code is hosted in a public GitHub repository.Read 11 Oct 2026 · bayesflow.org

Backend requirement

BayesFlow requires a separately installed JAX, PyTorch, or TensorFlow backend to run.Read 11 Oct 2026 · bayesflow.org

Backends

BayesFlow uses Keras 3 and supports JAX, PyTorch, and TensorFlow backends.Read 11 Oct 2026 · bayesflow.org

Bayesian analysis

YesRead 23 Sep 2026 · bayesflow.org

Deep learning backends

BayesFlow supports Keras 3 with JAX, PyTorch, or TensorFlow backends.Read 11 Oct 2026 · bayesflow.org

Diagnostics

Built-in diagnostics include calibration, parameter recovery, posterior contraction, and posterior predictive checks.Read 11 Oct 2026 · bayesflow.org

Inference workflows

It provides workflows for posterior, likelihood, ratio, point, and model comparison estimation.Read 11 Oct 2026 · bayesflow.org

Installation

The documentation provides pip and source installation instructions and says a machine-learning backend must also be installed.Read 11 Oct 2026 · bayesflow.org

Integration: Mamba

An optional Mamba summary network is backed by the mamba-ssm library.Read 11 Oct 2026 · bayesflow.org

Integration: PyMC

The optional PyMC wrapper lets trained ratio or continuous approximators act as likelihood terms in PyMC models.Read 11 Oct 2026 · bayesflow.org

Intended audience

The package metadata lists developers and science or research as intended audiences.Read 11 Oct 2026 · pypi.org

Limit

BayesFlow will not run without an installed machine-learning backend.Read 11 Oct 2026 · bayesflow.org

Maintainers

The project is managed by researchers from Rensselaer Polytechnic Institute, TU Dortmund University, and Heidelberg University.Read 11 Oct 2026 · bayesflow.org

Modularity

The simulator, adapter, summary network, inference network, and training strategy can be swapped independently.Read 11 Oct 2026 · bayesflow.org

Neural models

Its listed generative models include diffusion and consistency models.Read 11 Oct 2026 · github.com

PyMC limitation

The documentation says its PyMC integration is currently supported only with the JAX backend.Read 11 Oct 2026 · bayesflow.org

Python versions

The current installation instructions state support for Python 3.12 to 3.13.Read 11 Oct 2026 · github.com

Regression models

NoRead 23 Sep 2026 · bayesflow.org

Scripting support

YesRead 23 Sep 2026 · bayesflow.org

Statistical tests

NoRead 23 Sep 2026 · bayesflow.org

Workflow

Its workflow connects generative models, data adapters, neural inference, training, and simulation-based diagnostics.Read 11 Oct 2026 · bayesflow.org

Where Specifiction places it

• Statistical Analysis Software#61 of 1067.6