libbi ☆
« Back to VersTracker
Description:
Bayesian state-space modelling on parallel computer hardware
Type: Formula  |  Tracked Since: Dec 28, 2025
Links: Homepage  |  formulae.brew.sh
Category: Other
Tags: bayesian statistics modelling parallel-computing gpu
Install: brew install libbi
About:
Libbi is a Bayesian state-space modelling framework designed for high-performance computing. It leverages parallel hardware, including GPUs and clusters, to perform efficient particle MCMC and SMC^2 inference. Its primary value is enabling scalable analysis of complex, non-linear models commonly used in epidemiology, ecology, and finance.
Key Features:
  • Parallelized Particle MCMC and SMC^2 sampling
  • GPU acceleration support via OpenCL
  • Interoperability with R and Python for model control and analysis
  • Scalable across distributed computing clusters
Use Cases:
  • Epidemiological modelling for tracking disease spread
  • Ecological population dynamics analysis
  • Financial time-series forecasting
Alternatives:
  • Stan – Stan uses HMC/NUTS on CPU, whereas Libbi specializes in particle methods and GPU acceleration.
  • NIMBLE – NIMBLE is an R package offering flexible MCMC, while Libbi focuses specifically on particle filtering for state-space models.
Version History
Detected Version Rev Change Commit
Dec 29, 2025 3:55pm 6 REVISION_ONLY 6785e12c
Sep 14, 2025 2:51am 5 VERSION_BUMP 4ee5816f
Sep 13, 2024 1:52pm 5 VERSION_BUMP 9ff6ea99
« Back to VersTracker  |  Browse Packages  |  About