Thermodynamic Hypergraphical Model Library in JAX
-
Updated
Nov 10, 2026 - Python
Thermodynamic Hypergraphical Model Library in JAX
JAX Framework for Parameterized Stochastic Circuits (PSCs) and Directed Factor Graphs (DFGs) of Kernels
Software platform for Thermodynamic Sampling Units - probabilistic computing emulator
The world's first probabilistic processor. Distributions are the data type. Bayesian inference is a machine instruction. 2,237 SKY130 gates.
pymtinv is a Python-based scientific computing library designed for 2D Magnetotelluric (MT) inversion. It serves as a comprehensive research framework to benchmark classical deterministic optimization methods against novel Probabilistic Computing (P-bit) architectures tailored for FPGA acceleration.
Implementation of the core Adaptive Chirplet Transform (ACT) algorithm using THRML (Thermodynamic sampling) to efficiently find the best matching atom in the continuous space
Solving a NP-Complete problem (Max-Cut) with Extropic "Thermodynamic" sampling hardware.
This repo is an educational POC showing how thermodynamic sampling units can produce sparse polynomial approximations. It’s intended for demonstration, not production.
Thermodynamic computing in pure Rust: energy-based models, block-Gibbs, parallel tempering, thermodynamic linear algebra, stochastic differentiable programs, a variational compiler onto device topologies, and a joules ledger. Zero deps, wasm-clean, verified against exact physics — and it reaches real Ising silicon through the same trait as a CPU.
Stochastic spiking neurons that sample from Boltzmann distributions, with the cost of every sample. Thermodynamic computing meets neuromorphic engineering.
Encode and sample probabilistic weights in Python. NumPy-first, with optional THRML.
Extropic — independent third-party profile of a public API surface, by API Evangelist. Extropic is a hardware company building thermodynamic (probabilistic) computing systems designed to run generative AI workloads at dramatically lower energy cost than GPUs. Its core hardware is the Thermodynamic Sampling Unit (TSU) — an inherently probabilistic c
probbit: a personality, mood and memory layer for AI agents, on an exact p-bit decision engine. Every stance and plan is computed on a CPU, keeps its rules by construction, and is replayable and verifiable. CLI, MCP server, OpenClaw plugin, npm, Python, wasm. Zero-dependency Rust.
This repository reproduces the numerical results of the publication: Kaiser et al., "Subnanosecond fluctuations in low-barrier nanomagnets." Physical Review Applied 12, no. 5 (2019): 054056.
p-bit (probabilistic bit) image recognition and QUBO solver in C++
A Rust implementation of stochastic hashing (stochash), a probabilistic primitive whose input is a buffer of bytes (each interpreted as a probability byte/256) and whose output is a bit-string of the same length, with each bit sampled independently with P(bit_i = 1) = byte_i / 256.
A computer whose bits hold real numbers and are redrawn from an energy. Continuous probabilistic computing in numpy (arXiv:2610.31339).
The research behind PBit1: paper, plain-language overview, benchmarks, failure analysis, and reproducible artifacts.
Interactive 16 p-bit Langevin dynamics simulation in JAX over a POMDP, realizing autonomous perception, planning, and action on continuous energy landscapes.
Tokamak plasma tomography as Ising/Potts sampling with thrml — bounded-degree I-chain formulation and energy/latency estimates for Extropic's thermodynamic sampling unit
To associate your repository with the probabilistic-computing topic, visit your repo's landing page and select "manage topics."