4. Methodology Overview
How the methodology works at a high level.
Greenpixie's AI token methodology estimates the energy consumption and embodied emissions of cloud AI inference on a per-token basis. A representative set of open-source models is benchmarked on real GPU hardware, and regression models trained on those benchmarks generalise the findings across the full range of models in Greenpixie's database, including proprietary models that cannot be measured directly. Predicted energy values are then mapped to each cloud provider's billing configurations and combined with regional carbon intensity and hardware embodied emissions data to produce per-token carbon estimates.
The methodology operates through four components, being physical benchmarking, predictive modelling, billing configuration mapping, and carbon and embodied emissions calculation. Section 7 describes each in detail. Together they produce the per-token energy and carbon values applied to customer usage data within Greenpixie's data product.