LLM landscape
Compute vs capability
Track how published training compute relates to model capability in one dataset. Each point carries a status and source status so measured rows and explicit assumptions stay inspectable without splitting the view.
Show metrics
- Assumptions
- 14
- Review queue
- 108
Compute vs capability
Training compute vs capability
Nov 10, 2026, 5:20 AM
59 models place their published training compute against capability. Across 45 measured models, capability climbs about +12.5 ECI points per 10× of training compute. GPT-6 Astra sits at the capability frontier (169 ECI).
View and filters
Each dot is a model · log compute axis · higher and right-er means more capability per unit of training compute
Training compute is Epoch's published FLOP estimate on a log axis; capability is the ECI score. Points are coloured by organisation and shaped by weights access (square = open weights, circle = proprietary); outlined markers carry an assumed rather than measured value. The dashed trend line is a log-linear fit over visible measured rows only, so assumption rows plot but do not bend it. Hover near a point to inspect it, click to pin, Escape to clear, use legend filters, and Frontier focus to zoom crowded high-capability clusters.