Skip to content

Evaluation

MATH (Hendrycks; Gemini 1.5)

Hendrycks MATH benchmark; 1.5 Pro +14.5% over 1.0 Ultra; Flash +22.3% over 1.0 Pro

Core capabilities Google DeepMind — Gemini

How it was run

Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.

Gemini 1.5
Google DeepMind — Gemini

partial detail 7 method fields published

See how this test was run