Compare Models
Select models (max 5)
Gemini 3.8 FlashGPT-6 Sol
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 54.83%±1.03 | 57.54%±1.01 |
| Legal Research BenchAgentic US legal research | 38.94%±3.39 | 28.85%±3.15 |
| Finance Agent (v2)Core financial analyst tasks | 61.44%±0.13 | 49.05%±0.58 |
| Tax Agent BenchAgentic US corporate tax research | 66.77%±3.15 | 53.05%±3.29 |
| MedCodeMedical billing code support | 48.13%±2.18 | 47.07%±2.12 |
| Terminal-Bench ScienceExpert-authored scientific research workflows | 8.57%±3.37 | 30.00%±5.52 |
| Code MigrationRewriting programs in new languages | 36.55%±4.18 | 57.20%±4.21 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 19.19%±2.52 | 44.44%±3.54 |
| Vibe Code Bench v1.1Building web apps from scratch | 78.65%±3.88 | 87.82%±2.53 |
Benchmarks
Vals Index *
Gemini 3.8 Flash
0.00%± 1.03
(43/43)GPT-6 Sol
0.00%± 1.01
(43/43)Legal Research Bench *
Gemini 3.8 Flash
0.00%± 3.39
(73/73)GPT-6 Sol
0.00%± 3.15
(73/73)Finance Agent (v2) *
Gemini 3.8 Flash
0.00%± 0.13
(74/74)GPT-6 Sol
0.00%± 0.58
(74/74)Tax Agent Bench *
Gemini 3.8 Flash
0.00%± 3.15
(65/65)GPT-6 Sol
0.00%± 3.29
(65/65)MedCode *
Gemini 3.8 Flash
0.00%± 2.18
(104/104)GPT-6 Sol
0.00%± 2.12
(104/104)Terminal-Bench Science
Gemini 3.8 Flash
0.00%± 3.37
(38/38)GPT-6 Sol
0.00%± 5.52
(38/38)Code Migration *
Gemini 3.8 Flash
0.00%± 4.18
(73/73)GPT-6 Sol
0.00%± 4.21
(73/73)Terminal-Bench 4.0
Gemini 3.8 Flash
0.00%± 2.52
(43/43)GPT-6 Sol
0.00%± 3.54
(43/43)Vibe Code Bench v1.1 *
Gemini 3.8 Flash
0.00%± 3.88
(108/108)GPT-6 Sol
0.00%± 2.53
(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
45.31%
vs.15.26%
Finance
68.04%
vs.57.88%
Healthcare
66.32%
vs.64.55%
Math
48.00%
vs.83.00%
Science
35.76%
vs.45.00%
Academic
91.25%
vs.N/A
Education
35.06%
vs.44.79%
Coding
48.73%
vs.54.81%
Cyber
43.75%
vs.77.98%
Social Mobility
65.29%
vs.56.63%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$5.729
$7.581
Cost / TestVals Index
$5.729
vs.$7.581
Input Cost/ 1M Tokens
$1.50
vs.$2.00
Input Cache Write/ 1M Tokens
N/A
vs.$2.50
Input Cache Read/ 1M Tokens
$0.15
vs.$0.20
Output Cost/ 1M Tokens
$7.50
vs.$10.00
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Google | OpenAI |
| Latency | 3117.16s | 1750.68s |
| Cost (In/Out) | $1.5 / $7.5 | $2 / $10 |
| Context Window | 1M | 1M |
| Max Output Token | 66k | 128k |
| Input Modality |