benchgap
Calibration

IMOAnswerBench → HMMT Feb 2026

HMMT Feb 2026 is estimated from IMOAnswerBench with a linear curve fitted on 8 models measured on both: y = 1.1858·x + -0.1048, R² = 0.91, cross-validated error 9.7 pp. It is used for 1 estimate.

Estimated modelIMOAnswerBenchHMMT Feb 2026Source
dots3-note Preview90.9%97.3%estimated ± 9.7 pp, low confidence