Research: Energy Calculus
For quite some time, we have handled energy and power as physical quantities spent by running computation, and over time we developed mental models on how...
Research: Energy-Efficient Large-Model Training with Kareus
Our latest work on energy-efficient large-model training, Kareus, has been published at OSDI ’26! Energy has become a key bottleneck to AI scaling. Most...
Release: The ML.ENERGY Data Toolkit
Programmatic access to ML.ENERGY Benchmark data.
Engineering: Thermally Stable GPU Energy Profiling
Accurate GPU energy measurement is harder than it looks, especially when profiling short workloads or running trials back-to-back. In our new technical blog...
Analysis: Energy Improvements Over Time
A year's progress on energy efficiency with software optimizations.
Release: The ML.ENERGY Leaderboard v3.0
GPT OSS, DeepSeek, reasoning, B200, and more.