Freedom251 › Statistical Analysis Software

greta

Free plan

What greta is

greta is free desktop software for writing statistical models in R and fitting them with Markov chain Monte Carlo. Its interactive workflow lets users build greta arrays in R and combine them with mathematical operators and functions. The package supports Bayesian analysis with prior distributions and frequentist inference using variables without probability distributions, plus regression models and scripting. It accepts R numeric or logical vectors, matrices, arrays, and numeric or logical data frames. Users can extend projects with their own R functions or packages. A plotting function helps visualize and check models before sampling. greta uses Google TensorFlow and TensorFlow Probability and can run on CPU clusters and GPUs. The stable package is on CRAN, while development versions are available through r-universe and GitHub. On first use, compatible Python, TensorFlow, and TensorFlow Probability are usually installed through uv; conda setup is documented for offline or pinned environments. Model plotting additionally needs DiagrammeR and igraph.

Who it suits

It suits R users who want Bayesian or frequentist modeling with MCMC without learning BUGS or Stan. It also fits users needing regression, scripting, or extensible R workflows.

What it does well

Free Apache License 2.0 package
Bayesian and frequentist inference
Runs on CPUs clusters and GPUs
Interactive modeling directly in R
Supports custom R functions and packages

What to know first

Plotting needs DiagrammeR and igraph
Python and TensorFlow dependencies are installed
Usage questions go to the greta forum
Plotting is unavailable without extra packages

Verdict

greta provides an R-centered route to Bayesian and frequentist MCMC modeling, with TensorFlow-backed computation and optional visualization. Plan for its Python dependencies and separate plotting packages when setting it up.

Facts from greta's own pages

Alternative dependency setup: The maker documents using a conda environment when needed, including for offline use or pinned versions. · greta-dev.github.io, 11 Oct 2026
Analysis types: The documentation describes Bayesian analyses with priors and frequentist inference using variables without probability distributions. · greta-dev.github.io, 11 Oct 2026
Bayesian analysis: Yes · greta-dev.github.io, 23 Sept 2026
Bayesian and frequentist modeling: greta supports Bayesian analyses with prior distributions and frequentist inference using variables without probability distributions. · greta-dev.github.io, 11 Oct 2026
Compute: The maker says greta can run on CPU clusters and GPUs. · greta-dev.github.io, 11 Oct 2026
Data import formats: R numeric/logical vectors, matrices, arrays, and numeric/logical data frames · greta-dev.github.io, 23 Sept 2026
Dependencies: greta uses Google's TensorFlow and TensorFlow Probability Python packages, which it usually installs automatically on first use via uv. · greta-dev.github.io, 11 Oct 2026
Dependency setup: On first use, greta usually installs a compatible Python, TensorFlow, and TensorFlow Probability through uv automatically. · greta-dev.github.io, 11 Oct 2026
Deployment: desktop · greta-dev.github.io, 23 Sept 2026
Developers: The site credits Nick Golding and Nicholas Tierney as developers. · greta-dev.github.io, 11 Oct 2026
Extensibility: Users can write their own R functions and packages using greta. · greta-dev.github.io, 11 Oct 2026
Free plan: Yes · greta-dev.github.io, 23 Sept 2026
Inference: greta builds statistical models interactively in R and samples from them using MCMC. · greta-dev.github.io, 11 Oct 2026
Installation: The stable package is available from CRAN, and development versions are available through r-universe and GitHub. · greta-dev.github.io, 11 Oct 2026
License: The CRAN package listing identifies greta's license as Apache License 2.0. · cran.mirror.garr.it, 11 Oct 2026
Modeling: greta models are written in R, so users do not need to learn another language such as BUGS or Stan. · greta-dev.github.io, 11 Oct 2026
Model visualization: greta provides a plotting function to visualize and check models before sampling. · greta-dev.github.io, 11 Oct 2026
Optional plotting dependency: DiagrammeR is not installed automatically, and users need it with igraph to plot greta models. · greta-dev.github.io, 11 Oct 2026
Plotting limitation: If users cannot install igraph and DiagrammeR, they can still use greta except for plotting models. · greta-dev.github.io, 11 Oct 2026
Plotting requirement: Model plotting depends on DiagrammeR, which is not installed automatically with greta. · greta-dev.github.io, 11 Oct 2026
Purpose: greta lets users write statistical models in R and fit them by MCMC. · greta-dev.github.io, 11 Oct 2026
Regression models: Yes · greta-dev.github.io, 23 Sept 2026
R workflow: Models are written interactively in R, where greta arrays can be combined with mathematical operators and functions. · greta-dev.github.io, 11 Oct 2026
Scripting support: Yes · greta-dev.github.io, 23 Sept 2026

Where greta stands on Freedom251

Statistical Analysis Software#60 of 107