Gemini 3.1 Pro Preview (02/26)

Release Date: 2/19/2026

Vals Index

Accuracy (Vals Index)

53.77% ± 1.59

Latency (Vals Index)

435.11s

Cost/Test (Vals Index)

$1.35

Context Window

1M

Max Output Tokens

66k

Input Modality

Hyperparameter settings
Default Provider : Google

Temperature

1

Top P

Default

Top K

Default

Max Output Tokens

65,536

Reasoning Effort

high

Benchmarks

Accuracy

Rankings

0.0%

± 1.59
12/ 30

0.0%

± 1.41
9/ 21

0.0%

± 3.93
15/ 23

0.0%

± 0.94
25/ 116

0.0%

± 1.21
13/ 28

0.0%

± 2.81
10/ 13

0.0%

± 2.00
1/ 68

0.0%

± 1.92
40/ 65

0.0%

± 0.91
3/ 80

0.0%

± 4.39
12/ 42

0.0%

± 3.29
16/ 61

0.0%

± 1.30
10/ 12

0.0%

± 0.86
41/ 122

0.0%

± 4.34
29/ 66

0.0%

± 1.05
1/ 116

0.0%

± 0.93
2/ 122

0.0%

± 0.33
2/ 119

0.0%

± 0.28
2/ 115

0.0%

± 0.78
4/ 76

0.0%

± 0.00
12/ 24

0.0%

± 1.83
9/ 64

0.0%

± 1.12
5/ 35
Contact us
Or send us an email at contact@vals.ai

License type:

Proprietary (contact us to get access)
Industry Partner
Academic

Read our methodology.