Compare Models
Select models (max 5)
DeepSeek V4.1 FlashGrok 4.6
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 51.32%±1.13 | 52.09%±1.15 |
| Legal Research BenchAgentic US legal research | 41.35%±3.42 | 48.08%±3.47 |
| Finance Agent (v2)Core financial analyst tasks | 53.48%±0.39 | 53.68%±0.67 |
| Tax Agent BenchAgentic US corporate tax research | 62.46%±3.15 | 70.79%±3.00 |
| MedCodeMedical billing code support | 41.17%±2.04 | 44.71%±2.26 |
| Terminal-Bench ScienceExpert-authored scientific research workflows | 0.00%±0.00 | 11.43%±3.83 |
| Code MigrationRewriting programs in new languages | 45.62%±4.29 | 44.57%±4.48 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 19.70%±1.75 | 17.17%±1.34 |
| Vibe Code Bench v1.1Building web apps from scratch | 84.74%±2.89 | 76.24%±3.82 |
Benchmarks
Vals Index *
DeepSeek V4.1 Flash
0.00%± 1.13
(43/43)Grok 4.6
0.00%± 1.15
(43/43)Legal Research Bench *
DeepSeek V4.1 Flash
0.00%± 3.42
(73/73)Grok 4.6
0.00%± 3.47
(73/73)Finance Agent (v2) *
DeepSeek V4.1 Flash
0.00%± 0.39
(74/74)Grok 4.6
0.00%± 0.67
(74/74)Tax Agent Bench *
DeepSeek V4.1 Flash
0.00%± 3.15
(65/65)Grok 4.6
0.00%± 3.00
(65/65)MedCode *
DeepSeek V4.1 Flash
0.00%± 2.04
(104/104)Grok 4.6
0.00%± 2.26
(104/104)Terminal-Bench Science
DeepSeek V4.1 Flash
0.00%± 0.00
(38/38)Grok 4.6
0.00%± 3.83
(38/38)Code Migration *
DeepSeek V4.1 Flash
0.00%± 4.29
(73/73)Grok 4.6
0.00%± 4.48
(73/73)Terminal-Bench 4.0
DeepSeek V4.1 Flash
0.00%± 1.75
(43/43)Grok 4.6
0.00%± 1.34
(43/43)Vibe Code Bench v1.1 *
DeepSeek V4.1 Flash
0.00%± 2.89
(108/108)Grok 4.6
0.00%± 3.82
(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.50.07%
Finance
57.72%
vs.64.50%
Healthcare
63.34%
vs.65.62%
Math
54.00%
vs.51.00%
Science
29.65%
vs.38.09%
Academic
N/A
vs.92.05%
Education
47.88%
vs.28.90%
Coding
39.57%
vs.48.99%
Cyber
37.23%
vs.67.68%
Social Mobility
64.28%
vs.66.85%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$0.332
$4.493
Cost / TestVals Index
$0.332
vs.$4.493
Input Cost/ 1M Tokens
$0.30
vs.$2.00
Input Cache Read/ 1M Tokens
$0.006
vs.$0.50
Output Cost/ 1M Tokens
$1.20
vs.$6.00
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | DeepSeek | SpaceXAI |
| Latency | 1673.17s | 2221.07s |
| Cost (In/Out) | $0.3 / $1.2 | $2 / $6 |
| Context Window | 1M | 500k |
| Max Output Token | 384k | 500k |
| Input Modality |