Inference
EN
Epoch AI

OEIS Open

Evaluated by Epoch AI on their own harness.

ModelsModels with a result.3
Top resultBest result on this benchmark.29.9%Claude Opus 4.8
Top-3 spreadPoints from first to third.7.7 pts
YearYear of release.—

Results

3 results
#ModelResultThe score from the source.
01Claude Opus 4.8high effort29.9% ±2.1
02GPT-5.5medium effort26.2% ±2.0
03Gemini 3.5 Flashhigh effort22.2% ±1.9