thrml is an open-source library by Extropic AI for thermodynamic computing and probabilistic machine learning.
At a Glance
Pricing
Open Source
Fully open-source library available at no cost on GitHub.
Engagement
Available On
Web
API
SDK
Listed Mar 2026
About thrml
thrml is an open-source library developed by Extropic AI, hosted on GitHub, focused on thermodynamic computing and probabilistic machine learning primitives. It provides researchers and developers with tools to explore and implement thermodynamic approaches to AI and machine learning. The project is in early/active development and is publicly available for contribution and experimentation.
- Open-source codebase — Clone or fork the repository directly from GitHub to get started with thermodynamic ML primitives.
- Thermodynamic computing focus — Implements concepts from thermodynamic computing, enabling novel probabilistic inference and sampling approaches.
- Research-oriented — Designed for researchers and engineers exploring next-generation hardware-software co-design for AI.
- GitHub-hosted — All source code, issues, and contributions are managed through the GitHub repository at extropic-ai/thrml.
- Python-based — Built for the Python ecosystem, making it accessible to the broader ML research community.
Community Discussions
Be the first to start a conversation about thrml
Share your experience with thrml, ask questions, or help others learn from your insights.
Pricing
OPEN SOURCE
Open Source
Fully open-source library available at no cost on GitHub.
- Full source code access
- Community contributions
- GitHub issues and discussions
Capabilities
Key Features
- Thermodynamic computing primitives
- Probabilistic machine learning tools
- Open-source and community-driven
- Python-based library
- Research-oriented design
Integrations
Python
GitHub
API Available
View Docs