Extropic is a hardware startup founded in 2022 to develop energy-efficient computers using thermodynamic computing - a novel approach that harnesses thermal noise and probabilistic computation to process artificial intelligence workloads. The company has raised $14.1 million in seed funding from Kindred Ventures and Valor Equity Partners, with advisors including Naval Ravikant and Balaji Srinivasan.
The company's core innovation is thermodynamic sampling units (TSUs), hardware that directly samples from complex probability distributions rather than relying on traditional matrix multiplication. This approach exploits inherent randomness to achieve orders of magnitude better energy efficiency than conventional GPUs. Extropic's product lineup includes the XTR-0 development platform, the X0 and Z1 chips, and THRML, an open-source Python library that enables developers to simulate thermodynamic algorithms.
The founding team and wider staff bring experience from major technology firms including Google, IBM, Apple, and Microsoft, with backgrounds spanning machine learning and computer hardware. The company is based in the United States and is actively developing both proprietary hardware and supporting software infrastructure.
