Gemini 2.5
Google DeepMind — Gemini
rich detail 26 method fields published
Evaluation
RE-Bench (ML R&D acceleration)
Open-source Research Engineering Benchmark (Wijk et al., 2024/2025): seven ML challenges taking human practitioners hours; two omitted (internet access disallowed); METR methodology scoring; used for ML R&D Autonomy/Uplift Level 1 CCL assessment
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.