 | The topic of this articlemay not meet Wikipedia'sgeneral notability guideline. Please help to demonstrate the notability of the topic by citingreliable secondary sources that areindependent of the topic and provide significant coverage of it beyond a mere trivial mention. If notability cannot be shown, the article is likely to bemerged,redirected, ordeleted. Find sources: "Thermodynamic computing" – news ·newspapers ·books ·scholar ·JSTOR(November 2025) (Learn how and when to remove this message) |
Thermodynamic computing refers to a new type of computing that has reached latter stages of development as of 2025. This type of computing has been pioneered and developed by the computing companyExtropic.[1]
Stochastic computing was investigated as early as the 1960s and 1970s, when engineers proposed circuits that performedstochastic sampling rather than fixedBoolean logic.Boltzmann machines based onstatistical mechanics and energy-basedneural networks provided the theoretical foundation for using physical energy landscapes to represent probability distributions. This was also developed further inmachine-learning research on diffusion and generative models.
In the 2000s and 2010s, developments inquantum annealing, notablyD-Wave Systems computers andmemristive systems, further demonstrated how physical systems could relax toward low-energy states corresponding to computational solutions. Extropic's approach represents a continuation of this tradition, replacing fully digital logic with thermodynamic sampling units (TSUs) designed to exploit controlled fluctuations for energy-efficient inference.
Extropic developed a new type of computing hardware, the thermodynamic sampling unit (TSU). TSUs operate differently than conventional CPUs; instead of processing a series of programmable deterministic computations, TSUs produce samples from a programmable distribution.[1]
Entropic's hardware directly samples from complex probability distributions, omitting matrix multiplication TSUs sample from energy-based models (EBM), a type of machine learning model that directly define the shape of a probability distribution via an energy function.[1] This distinguishes them from conventional AI algorithms that are based on sampling from complex probability distribution; current AI systems generally produce a vector of probabilities, and then derive a sample from that.
The inputs to a TSU are parameters that specify the energy function of an EBM, and the outputs of a TSU are samples from the defined EBM. To use a TSU for machine learning, the parameters of the energy function are adjusted so that the EBM on the actual TSU will constitute a reliable model of real-world conditions.[1]
On October 29, 2025, Extropic announced the development of its very first hardware item, referred to as the Experimental Testing & Research Platform 0 (XTR-0).[2]
One article notes:
Normal Computing has announced its successful tape-out of the world's first thermodynamic computing chip, called CN101. Designed for AI/HPC data centers, the ASIC is a step away from traditional silicon computation methods that uses thermodynamics (and other physics principles) to reach computational efficiency that traditional chips can't match.
Thermodynamic chips are a world apart from traditional computing — closer in practice to the realms of quantum and probabilistic computing. Where noise is the enemy of standard electronics, thermodynamic and probabilistic chips actively use noise to solve problems.
“We’re focusing on algorithms that are able to leverage noise, stochasticity, and nondeterminism,” said Zachary Belateche, silicon engineering lead at Normal Computing, in a recent interview with IEEE Spectrum. “That algorithm space turns out to be huge, everything from scientific computing to AI to linear algebra."[3]
![[icon]](/mt/?noimg=&dark=on&url=https%3a%2f%2fen.m.wikipedia.org%2fwiki%2fFile%3aWiki_letter_w_cropped.svg) | This section is empty. You can help byadding to it.(October 2025) |
| This section is empty. You can help byadding to it.(October 2025) |
