Gemini 1.5
Google DeepMind — Gemini
rich detail 9 method fields published
Evaluation
MGSM (Gemini 1.5)
MGSM: 11 languages; 1.5 Pro improves almost +9% over 1.0 Ultra; Flash surpasses by >3
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.