readable AI benchmarks

Google

Gemini 3.1 Pro Preview

reasoning · closed weights · Feb 19, 2026

Quality Score64.2
Value Score42.0
Reliability65.9
Cache Discount90%

Model specification

Reasoning
reasoning
Input modalities
text, image, video, speech
Output modalities
text
Context window
1M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$2.00
Output
$12.00
Cached Input
$0.200
Cached Output
not available
Cost per task
$0.335

Capability

Intelligence
47.7
Coding
68.8
Agentic
23.0
Omniscience
31.9
Correct
54.9%
Blended price
$1.74

Answer outcomes

Correct 54.9%Incorrect 23.0%Abstained 22.2%

Artificial Analysis benchmarks

GDPval-AA v2
23%
τ³-Banking
21%
Terminal-Bench v2.1
74%
SciCode
59%
Humanity’s Last Exam
47%
GPQA Diamond
94%
CritPt
18%
AA-Omniscience
66%
AA-LCR
79%

Similar models