
Thermodynamic Computing: From 0 to 1 (Launch Video)
October 30th, 2025
原始链接: https://extropic.ai/
Extropic 正在开发革命性的热力学计算硬件,旨在比传统 GPU 更加节能,尤其是在 AI 工作负载方面。他们的核心创新——热力学采样单元 (TSU) 本质上是概率性的,与概率 AI 算法完美契合。 为了方便开发,Extropic 提供 XTR-0 平台,用于低延迟的芯片到处理器的通信,以及 THRML,一个用于算法创建和 TSU 模拟的开源 Python 库。 Extropic 通过多次媒体露面(Wired、TED AI、Lex Fridman & Garry Tan 的播客)和详细介绍其技术的出版物而备受关注。他们正在积极扩大团队,寻找工程师和科学家来推进这种新的计算范式。最终,Extropic 旨在以其节能的 AI 处理方法颠覆当前的数据中心格局。
Extropic is building thermodynamic computing hardware that is radically more energy efficient than GPUs.
Our thermodynamic sampling units (TSUs) are inherently probabilistic, the perfect fit for probabilistic AI workloads.
Hardware
prototype platform
XTR-0 enables the development of ultra-efficient AI algorithms by providing low-latency communication between Extropic chips and a traditional processor.

Software
Our open-source Python library that enables everyone to develop thermodynamic algorithms and simulate running them on TSUs

We are hiring engineers and scientists to help us pioneer a new form of computing.
