Back to models
Intelligence Index
Reasoning setting
29.7#34 of 83measured models
Measured · Artificial Analysis v4.3.2- API price from
- $2.00 in / $12.00 out / 1M tokOpenRouter · Google AI Studio
- Context window
- 1M tok
Performance profile
ReasoningRank #3 of 107
94.1%
GPQA DiamondCodingRank #12 of 94
58.7%
SciCodeCost efficiencyRank #55 of 67
22.9pts / $
Intelligence ÷ eval costCapabilities
Context window
Modalities
Measured setting
Integration
Reasoning trade-offs
Source (opens in a new tab)Gemini 3.1 Pro Preview
Cost per task (USD)
Gemini 3.1 Pro PreviewReasoning
- Intelligence Index
- 29.7
- Cost / task
- $1.30
- Evaluation time
- 2.1m
Pricing & access
USD per 1M tokens
- Input
- $2.00
- Output
- $12.00
- Cached input
- $0.20
Benchmarks
Full leaderboardSource (opens in a new tab)47.0%, rank 12 of 12658.7%, rank 12 of 944.0%, rank 33 of 8782.0%, rank 20 of 125792 Elo, rank 61 of 101
Humanity’s Last Exam
Frontier reasoning- 1Claude Opus 5.5Max61.4%
- 2Claude Fable 5.1Max59.1%
- 3Gemini 4 ArgonHigh57.1%
- 9MiMo-V2.6-ProReasoning49.4%
- 10Muse Spark 1.3Max48.7%
- 11Gemini 3.8 FlashHigh47.8%
- 12Gemini 3.1 Pro PreviewReasoning47.0%
- 13Kimi-K3Max46.9%
- 14Step 5 PreviewReasoning46.5%
- 15Claude Haiku 5.5Max44.4%
Peers use their best measured result on this benchmark.
Alternatives
- Scores higherBeam 1.0Reasoning
- Intelligence Index
- 30.3+0.6
- Cost / task
- $0.566−56%
- Higher score, lower costGPT-6.1 SolMax
- Intelligence Index
- 51.8+22.1
- Cost / task
- $0.724−44%
- More from Google
- Intelligence Index
- 22.2−7.6
- Cost / task
- $0.193−85%
Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.