Compare Models
Select models (max 5)
Gemini 3.1 Pro Preview (02/26)Gemini 4 Argon
 |  | |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 33.44%±1.13 | 68.90%±0.97 |
| Legal Research BenchAgentic US legal research | 20.67%±2.81 | 54.81%±3.46 |
| Finance Agent (v2)Core financial analyst tasks | 42.98%±1.21 | 65.40%±0.32 |
| Tax Agent BenchAgentic US corporate tax research | 37.84%±3.10 | 76.23%±2.87 |
| MedCodeMedical billing code support | 59.06%±2.00 | 58.80%±2.10 |
| Terminal-Bench ScienceExpert-authored scientific research workflows | 1.43%±1.43 | 44.29%±5.98 |
| Code MigrationRewriting programs in new languages | 17.31%±3.93 | 68.17%±4.35 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 2.52%±0.51 | 57.58%±2.31 |
| Vibe Code Bench v1.1Building web apps from scratch | 32.03%±4.34 | 91.91%±1.90 |
Benchmarks
Vals Index *
Gemini 3.1 Pro Preview (02/26)
0.00%± 1.13
(43/43)Gemini 4 Argon
0.00%± 0.97
(43/43)Legal Research Bench *
Gemini 3.1 Pro Preview (02/26)
0.00%± 2.81
(73/73)Gemini 4 Argon
0.00%± 3.46
(73/73)Finance Agent (v2) *
Gemini 3.1 Pro Preview (02/26)
0.00%± 1.21
(74/74)Gemini 4 Argon
0.00%± 0.32
(74/74)Tax Agent Bench *
Gemini 3.1 Pro Preview (02/26)
0.00%± 3.10
(65/65)Gemini 4 Argon
0.00%± 2.87
(65/65)MedCode *
Gemini 3.1 Pro Preview (02/26)
0.00%± 2.00
(104/104)Gemini 4 Argon
0.00%± 2.10
(104/104)Terminal-Bench Science
Gemini 3.1 Pro Preview (02/26)
0.00%± 1.43
(38/38)Gemini 4 Argon
0.00%± 5.98
(38/38)Code Migration *
Gemini 3.1 Pro Preview (02/26)
0.00%± 3.93
(73/73)Gemini 4 Argon
0.00%± 4.35
(73/73)Terminal-Bench 4.0
Gemini 3.1 Pro Preview (02/26)
0.00%± 0.51
(43/43)Gemini 4 Argon
0.00%± 2.31
(43/43)Vibe Code Bench v1.1 *
Gemini 3.1 Pro Preview (02/26)
0.00%± 4.34
(108/108)Gemini 4 Argon
0.00%± 1.90
(108/108)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-02
Comparison by Industry
Model performance on different sections of the economy.
Legal
36.02%
vs.54.23%
Finance
55.14%
vs.72.29%
Healthcare
67.59%
vs.73.11%
Math
26.00%
vs.99.00%
Science
27.38%
vs.55.36%
Academic
91.55%
vs.N/A
Education
48.68%
vs.53.65%
Coding
34.71%
vs.64.03%
Cyber
N/A
vs.61.07%
Social Mobility
53.79%
vs.69.76%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$2.18
$15.68
Cost / TestVals Index
$2.18
vs.$15.68
Input Cost/ 1M Tokens
$2.00
vs.$4.00
Input Cache Read/ 1M Tokens
$0.20
vs.$0.20
Output Cost/ 1M Tokens
$12.00
vs.$20.00
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Google | Google |
| Latency | 591.99s | 2792.77s |
| Cost (In/Out) | $2 / $12 | $4 / $20 |
| Context Window | 1M | 1M |
| Max Output Token | 66k | 262k |
| Input Modality |