anesthetic review
A focused Python package for inspecting and visualizing Bayesian sampling outputs.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
Researchers working with Bayesian sampling outputs can use anesthetic to process and visualize nested-sampling chains, with support for standard MCMC chains too. It is a Python package for local use, suited to analysts who want to examine sampling results and produce marginalized one- and two-dimensional plots. Its scope is specialized: the listed capabilities center on Bayesian sampling and chain analysis rather than general statistical testing, so users seeking a broader statistical analysis environment may prefer another tool.
The package’s analytical depth is its main distinction. It computes Bayesian evidences, Kullback–Leibler divergences, and Bayesian model dimensionalities, and supports kernel density estimation with boundary correction. Nested-sampling runs can be replayed dynamically or inspected through an interactive command-line view. Those features make anesthetic relevant when post-processing and visualization of sampling runs are central to a research workflow, rather than occasional tasks alongside a wide range of statistical methods.
For ecosystem fit, anesthetic reads supported chains from PolyChord, MultiNest, UltraNest, Nested_fit, CosmoMC, and Cobaya, as well as GetDist-compatible chains and its own CSV files. It can work with samples stored in pandas data structures. The project is free and open source, with documentation as its listed support channel. Analysts already working with these chain formats who need focused Bayesian post-processing are the clearest fit; teams that need broader statistical testing or a hosted subscription should look elsewhere.
anesthetic pros and cons
- Where it wins
- Computes Bayesian evidences, divergences, and model dimensionalities
- Replays nested-sampling runs and offers an interactive command-line view
- Reads chains from several samplers and supports KDE with boundary correction
- Where it doesn't
- Focused on Bayesian sampling outputs rather than general statistical testing
- Requires a local Python workflow rather than a hosted subscription
- Support is through documentation
anesthetic fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update