Compare Models
Select models (max 5)
Qwen 3.7 MaxDeepSeek V4
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | N/A | 38.63%±1.16 |
| Legal Research BenchAgentic US legal research | 25.48%±3.03 | 23.08%±2.93 |
| Finance Agent (v2)Core financial analyst tasks | 47.78%±0.73 | 44.08%±0.65 |
| Tax Agent BenchAgentic US corporate tax research | 47.41%±3.24 | 58.50%±3.13 |
| MedCodeMedical billing code support | 38.75%±2.20 | 40.45%±2.12 |
| Code MigrationRewriting programs in new languages | 13.07%±2.86 | 26.20%±4.04 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | N/A | 11.11%±1.34 |
| Vibe Code Bench v1.1Building web apps from scratch | 47.67%±4.63 | 49.93%±4.77 |
Benchmarks
Vals Index *
Qwen 3.7 Max
N/ADeepSeek V4
0.00%± 1.16
(43/43)Legal Research Bench *
Qwen 3.7 Max
0.00%± 3.03
(73/73)DeepSeek V4
0.00%± 2.93
(73/73)Finance Agent (v2) *
Qwen 3.7 Max
0.00%± 0.73
(74/74)DeepSeek V4
0.00%± 0.65
(74/74)Tax Agent Bench *
Qwen 3.7 Max
0.00%± 3.24
(65/65)DeepSeek V4
0.00%± 3.13
(65/65)MedCode *
Qwen 3.7 Max
0.00%± 2.20
(104/104)DeepSeek V4
0.00%± 2.12
(104/104)Code Migration *
Qwen 3.7 Max
0.00%± 2.86
(73/73)DeepSeek V4
0.00%± 4.04
(73/73)Terminal-Bench 4.0
Qwen 3.7 Max
N/ADeepSeek V4
0.00%± 1.34
(43/43)Vibe Code Bench v1.1 *
Qwen 3.7 Max
0.00%± 4.63
(108/108)DeepSeek V4
0.00%± 4.77
(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
37.35%
vs.35.72%
Finance
56.88%
vs.56.57%
Healthcare
59.07%
vs.57.80%
Math
N/A
vs.16.00%
Academic
89.73%
vs.88.32%
Coding
54.15%
vs.43.34%
Social Mobility
N/A
vs.62.92%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
N/A
$1.306
Cost / TestVals Index
N/A
vs.$1.306
Input Cost/ 1M Tokens
$2.50
vs.$1.32
Input Cache Write/ 1M Tokens
$3.125
vs.N/A
Input Cache Read/ 1M Tokens
$0.25
vs.$0.044
Output Cost/ 1M Tokens
$7.50
vs.$3.96
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Alibaba | DeepSeek |
| Latency | N/A | 1928.91s |
| Cost (In/Out) | $2.5 / $7.5 | $1.32 / $3.96 |
| Context Window | 984k | 1M |
| Max Output Token | 66k | 384k |
| Input Modality |