Google: Gemini 3.1 Pro Preview
Google: Gemini 3.1 Pro Preview is google's model. On the un-leaked meo benchmark it ranks #3 of 22 with 60.4% accuracy and an Effective Value (π) of 0.0038 β earning a correct answer for about $0.0968 in 69s each.
as of 2026-06-08
Per-domain accuracy
Efficiency
- Cost per correct
- $0.0968
- Seconds per correct
- 69
- Tokens per correct
- 7,867
- Correct answers
- 139 / 251
- Total benchmark cost
- $13.45
- Avg tokens / item
- 4,356
Robustness under pressure
Susceptibility index 0.021 (Ξaccuracy under threatening/emotional prompts; β0 = robust), sway magnitude 0.042. How we measure this β
- control
- 54%
- neutral
- 54%
- user direct
- 54%
- model direct
- 58%
- user implied
- 50%
- model implied
- 63%
- positive control
- 58%
Aggregated third-party benchmarks
Artificial Analysis (artificialanalysis.ai). Redistribution requires an AA commercial license.
Model details
- Lab
- Released
- 2026-02-19
- Weights
- β
- Context
- 1.0M
- Max output
- 66K
- Knowledge cutoff
- β
- Price in
- $2.00/M
- Price out
- $12.00/M
- Modality
- text+image+file+audio+video->text
Sources: openrouter Β· artificial_analysis Β· meo
FAQ
How accurate is Google: Gemini 3.1 Pro Preview?
Google: Gemini 3.1 Pro Preview scores 60.4% overall on the meo benchmark (rank #3 of 22), an un-leaked private holdout graded objective-first.
What is Google: Gemini 3.1 Pro Preview's Effective Value?
Its Effective Value (π) is 0.0038 at N=10, Ο=1, Ξ΄=1.5 β a fused metric of accuracy, speed, cost, and error-cascade. π rankings shift with task depth.
How much does Google: Gemini 3.1 Pro Preview cost per correct answer?
About $0.0968 per correct answer (7,867 tokens, 69 seconds each) at benchmark-time pricing via OpenRouter.
Is Google: Gemini 3.1 Pro Preview open weights?
Open-weights status for Google: Gemini 3.1 Pro Preview is not recorded.