Gemini 1.5
Google DeepMind — Gemini
rich detail 7 method fields published
Evaluation
HumanEval (Gemini 1.5)
Standard code-completion benchmark; leakage analysis: continued pre-training on test split boosted scores from 74.4% to 89.0%
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.