Compare Models
Select models (max 5)
DeepSeek V4.1 FlashGemini 3.7 Flash
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 51.32%±1.13 | 51.27%±1.10 |
| Legal Research BenchAgentic US legal research | 41.35%±3.42 | 34.62%±3.31 |
| Finance Agent (v2)Core financial analyst tasks | 53.48%±0.39 | 59.04%±0.27 |
| Tax Agent BenchAgentic US corporate tax research | 62.46%±3.15 | 57.66%±3.31 |
| MedCodeMedical billing code support | 41.17%±2.04 | 53.39%±2.12 |
| Terminal-Bench ScienceExpert-authored scientific research workflows | 0.00%±0.00 | 7.14%±3.10 |
| Code MigrationRewriting programs in new languages | 45.62%±4.29 | 34.80%±4.22 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 19.70%±1.75 | 12.12%±3.03 |
| Vibe Code Bench v1.1Building web apps from scratch | 84.74%±2.89 | 70.39%±4.84 |
Benchmarks
Vals Index *
DeepSeek V4.1 Flash
0.00%± 1.13
(43/43)Gemini 3.7 Flash
0.00%± 1.10
(43/43)Legal Research Bench *
DeepSeek V4.1 Flash
0.00%± 3.42
(73/73)Gemini 3.7 Flash
0.00%± 3.31
(73/73)Finance Agent (v2) *
DeepSeek V4.1 Flash
0.00%± 0.39
(74/74)Gemini 3.7 Flash
0.00%± 0.27
(74/74)Tax Agent Bench *
DeepSeek V4.1 Flash
0.00%± 3.15
(65/65)Gemini 3.7 Flash
0.00%± 3.31
(65/65)MedCode *
DeepSeek V4.1 Flash
0.00%± 2.04
(104/104)Gemini 3.7 Flash
0.00%± 2.12
(104/104)Terminal-Bench Science
DeepSeek V4.1 Flash
0.00%± 0.00
(38/38)Gemini 3.7 Flash
0.00%± 3.10
(38/38)Code Migration *
DeepSeek V4.1 Flash
0.00%± 4.29
(73/73)Gemini 3.7 Flash
0.00%± 4.22
(73/73)Terminal-Bench 4.0
DeepSeek V4.1 Flash
0.00%± 1.75
(43/43)Gemini 3.7 Flash
0.00%± 3.03
(43/43)Vibe Code Bench v1.1 *
DeepSeek V4.1 Flash
0.00%± 2.89
(108/108)Gemini 3.7 Flash
0.00%± 4.84
(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
43.76%
vs.43.54%
Finance
57.72%
vs.65.88%
Healthcare
63.34%
vs.68.67%
Math
54.00%
vs.58.00%
Science
29.65%
vs.7.14%
Academic
N/A
vs.91.01%
Education
47.88%
vs.49.23%
Coding
39.57%
vs.52.56%
Cyber
37.23%
vs.24.16%
Social Mobility
64.28%
vs.N/A
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$0.332
$4.709
Cost / TestVals Index
$0.332
vs.$4.709
Input Cost/ 1M Tokens
$0.30
vs.$1.50
Input Cache Read/ 1M Tokens
$0.006
vs.$0.15
Output Cost/ 1M Tokens
$1.20
vs.$7.50
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | DeepSeek | Google |
| Latency | 1673.17s | 2847.08s |
| Cost (In/Out) | $0.3 / $1.2 | $1.5 / $7.5 |
| Context Window | 1M | 1M |
| Max Output Token | 384k | 66k |
| Input Modality |