Artificial Analysis: Gemini Decision 1 scores 66 on its Decision Index, 7 points behind the leaders

Artificial Analysis says Google's Gemini Decision 1 scores 66 on its Decision Index v2.1, in line with Elephant 9B and Jev 1.14 and 7 points behind Drex Max v1.1 and GLiDE v1.2, which both score 73. It calls the model not yet frontier on the index.

• Strengths: it leads decision models on knowledge and reasoning, with 66 on AA-Omniscience-Decision and 49 on HLE-MultiChoice. Its Brier score (a calibration measure, lower is better) is 0.18 across the index.
• Weaknesses: on AA-Paperclip, its new classification and extraction benchmark, it answers all 15 questions correctly on 33% of documents, against 63% for Drex Max v1.1.
• Cost and speed: at $0.10 per 1M input tokens it costs $345 per 1M tasks, about three times Drex Max v1.1 ($103) and close to GPT-6 Luna in Decisions mode ($339). Time per task is 211 ms, faster than Drex Max v1.1 (283 ms) and slower than Celeris-1.1 Decision (61 ms).

These are Artificial Analysis's own measurements.