extropic-ai/thrml
A JAX library for GPU-accelerated block Gibbs sampling and energy-based probabilistic graphical models.
Not currently ranked — collecting fresh signals.
star history
THRML provides tools for building sparse, heterogeneous probabilistic graphical models with efficient block sampling. The library compiles factor-based interactions to compact state representations, minimizing Python loops and maximizing array-level parallelism in JAX. It enables experimentation with block Gibbs sampling and discrete energy-based models, targeting GPU acceleration for probabilistic inference tasks.
Frequently asked
- What is extropic-ai/thrml?
- A JAX library for GPU-accelerated block Gibbs sampling and energy-based probabilistic graphical models.
- Is thrml open source?
- Yes — extropic-ai/thrml is open source, released under the Apache-2.0 license.
- What language is thrml written in?
- extropic-ai/thrml is primarily written in Python.
- How popular is thrml?
- extropic-ai/thrml has 1.1k stars on GitHub.
- Where can I find thrml?
- extropic-ai/thrml is on GitHub at https://github.com/extropic-ai/thrml.