Epoch Capabilities Index | Epoch AI
epoch.ai · 334 words · saved by 1 readers
The Epoch Capabilities Index combines many benchmarks into a single capability scale for comparing models over time.
The general ECI is a composite metric which uses scores from over 50 distinct benchmarks to generate a single, general capability scale. At a high level, ECI stitches together its component benchmarks, determining their relative difficulty by making comparisons wherever models are evaluated on multiple benchmarks. Individual models obtain higher ECI scores if they perform better on harder benchmarks. We give an overview of our methodology here; further technical details are available in our paper, A Rosetta Stone for AI Benchmarks, which was funded by Google DeepMind, and written in…
saved by
related reading
- Benchmark Scores = General Capability + Claudinesssubstack.com
- Why benchmarking is hard | Epoch AIepoch.ai
- Benchmark Scores = General Capability + Claudinessepochai.substack.com
- AI Model & API Providers Analysis | Artificial Analysisartificialanalysis.ai
- Your Evals Will Break and You Won't See It Coming - Lun Wangwanglun1996.github.io
- The least understood driver of AI progress | Epoch AIepoch.ai
- AI progress is about to speed up | Epoch AIepoch.ai
- Are AI benchmarks doomed? - by Anson Ho and Greg Burnhamepochai.substack.com
- CAIS AI Dashboarddashboard.safe.ai
- Introducing the Conceptual Reasoning Indexalignment.anthropic.com
- Coding vs thinking — Paradigm 3paradigm3.org
- FrontierMath: Evaluating advanced mathematical reasoning in AI | Epoch AI | Epoch AIepochai.org