Gemini 1.0
Google DeepMind — Gemini
rich detail 10 method fields published
Evaluation
Complex prompts instruction-following evaluation (internal)
Fine-grained evaluation of complex prompts with multiple instructions: per-instruction accuracy and full-response accuracy on an internal dataset of prompts with varying complexity
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.