BayesFlow
bayesflow.org
- Free plan
- Windows
- macOS
- Linux
- Self-hosted
7.6 Club score Visit bayesflow.org
What the maker publishes
- Backend requirement
- BayesFlow requires a separately installed JAX, PyTorch, or TensorFlow backend to run. source
- Backends
- BayesFlow uses Keras 3 and supports JAX, PyTorch, and TensorFlow backends. source
- Bayesian analysis
- Yes source
- Deep learning backends
- BayesFlow supports Keras 3 with JAX, PyTorch, or TensorFlow backends. source
- Deployment
- desktop source
- Diagnostics
- The documentation demonstrates default diagnostics for inspecting inference results. source
- Diagnostics
- Built-in diagnostics include calibration, parameter recovery, posterior contraction, and posterior predictive checks. source
- Free plan
- Yes source
- Inference workflows
- It provides workflows for posterior, likelihood, ratio, point, and model comparison estimation. source
- Installation
- The project documents installation from PyPI with pip or uv, and installation from source for the latest features. source
- Installation
- The documentation provides pip and source installation instructions and says a machine-learning backend must also be installed. source
- Integration: Mamba
- An optional Mamba summary network is backed by the mamba-ssm library. source
- Integration: PyMC
- The optional PyMC wrapper lets trained ratio or continuous approximators act as likelihood terms in PyMC models. source
- Integrations
- Optional wrappers integrate Mamba summary networks through mamba-ssm and allow BayesFlow approximators to be used as likelihood terms in PyMC models. source
- Intended audience
- The package metadata lists developers and science or research as intended audiences. source
- License
- The package metadata identifies BayesFlow as MIT licensed. source
- License
- BayesFlow is released under the MIT License and its source code is hosted in a public GitHub repository. source
- Limit
- BayesFlow will not run without an installed machine-learning backend. source
- Maintainers
- The project is managed by researchers from Rensselaer Polytechnic Institute, TU Dortmund University, and Heidelberg University. source
- Modularity
- The simulator, adapter, summary network, inference network, and training strategy can be swapped independently. source
- Neural models
- Its listed generative models include diffusion and consistency models. source
- Product
- BayesFlow is a Python library for simulation-based amortized Bayesian inference with neural networks. source
- Purpose
- It trains neural networks on simulated datasets to produce Bayesian quantities such as posterior samples, likelihoods, likelihood ratios, point estimates, and model probabilities. source
- PyMC limitation
- The documentation says its PyMC integration is currently supported only with the JAX backend. source
Checked against the maker's pages on 11 Oct 2026.