[ Front page ] [ The edge ] [ Finished ] [ About ] [ Links ]
THIS ONE IS FINISHED. It stopped being a frontier on 2026-09-07: saturated. The page stays, because what it recorded still happened.
Above ninety-five per cent, which is the accuracy of the domain experts the test was calibrated against. It closed on the day this index opened.
95.8 per cent
Held by GPT-6 Astra. Set 2026-09-03. Written down here 2026-09-07.
Questions in biology, physics and chemistry written so that a specialist in a neighbouring field cannot answer them even with a search engine to hand. When it was published in 2023 it was expected to last for years. Models now score above ninety-five per cent, at or beyond the accuracy of the experts who wrote the questions, so it no longer separates anything from anything.
That description was written when this frontier was opened, on 2026-09-07, and is never rewritten. Where it and the number above it disagree, the number is the reading and the description is history.
Where to check it: https://epoch.ai/data/ai-benchmarking-dashboard
The program reads it from https://epoch.ai/data/benchmarks.csv, which is the same figure in a form a script can parse.
Previous: The largest quantum volume Next: Hard competition mathematics
[ Front page ] [ The edge ] [ Finished ] [ About ] [ Links ]