TuringLang/Turing.jl

Health v2 · maintenance only

Adoption not part of the score

2258 stars · 245 forks observed · 2026-10-01

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Turing.jl is a probabilistic programming library in Julia for Bayesian inference. Models are defined with an @model macro resembling mathematical notation, and inference is performed via MCMC samplers such as NUTS/HMC.

Use cases

• fit bayesian models with mcmc in julia
• perform hamiltonian monte carlo inference on probabilistic models
• define statistical models with a math-like dsl
• do bayesian linear regression with posterior sampling
• model hierarchical and probabilistic graphical models
• run bayesian neural network inference
• sample from posteriors with discrete and continuous parameters

When to choose

• you work in Julia and want expressive probabilistic programming with MCMC
• you need general-purpose Bayesian inference including discrete parameters and stochastic control flow
• you want model syntax close to mathematical specification
• you need interoperability with the Julia ecosystem and multiple AD backends

When to avoid

• you need a non-Julia language like Python or R (consider Stan, PyMC, or NumPyro)
• you need fast VI-only workflows or heavy GPU deep-learning training rather than Bayesian inference
• you require commercial-grade support with guaranteed response times

Facets

library · maturity active

machine-learning simulation data-science machine-learning artificial-intelligence data-science windows cross-platform probabilistic-programming bayesian-inference mcmc hamiltonian-monte-carlo bayesian-statistics statistical-modeling algorithms julia linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="TuringLang/Turing.jl")