AI Energy Monitoring
Carbontracker
Open-source tool that tracks and predicts energy use and carbon footprint of model training.
- Pricing
- Free
- Compliance
- Not stated
- Platforms
- On-premise, Cloud
AI Energy Monitoring
Cross-platform command-line tool that measures software energy consumption and writes it to CSV.
EnergiBridge is a command-line tool that wraps any command and records power, frequency, temperature, CPU/GPU usage and memory at fixed intervals to a CSV file through one interface, so energy experiments run and reproduce the same way across operating systems and CPUs. It covers Intel and AMD CPUs on Linux and Windows, M-series Macs, and NVIDIA GPUs via NVML, plus Intel, AMD and Apple GPUs on macOS. Measurements are written per interval in joules, MHz and volts, and a --summary flag prints total energy consumption at the end of a run. Prebuilt binaries are published on the releases page, and the tool can be built from source with a Rust toolchain.
Fits researchers and engineers who want a command-line wrapper that records power, frequency, temperature and usage series to CSV for any command, so energy experiments reproduce across Linux, Windows, macOS and different CPUs. Per-interval values in joules, MHz and volts plus a summary total make scripted comparisons straightforward. The trade-off is scope: it measures what the wrapped command draws on supported hardware and writes files, but converts nothing to carbon and has no API, dashboard or multi-node orchestration. Teams needing fleet-level monitoring or emissions reporting should pair it with other tooling.
Written by the GreenAIX editors from the recorded fields above. Vendors do not review or approve verdicts.
AI Energy Monitoring
Open-source tool that tracks and predicts energy use and carbon footprint of model training.
AI Energy Monitoring
Open-source Python library that estimates the carbon emissions of compute.
AI Energy Monitoring
Python library for tracking the CO2 emissions of machine learning models.