Compare Models
Select models (max 7)
DeepSeek V4Mercury 2.5
 |  |  |
|---|---|---|
| Legal Research BenchAgentic US legal research | 23.08%±2.93 | 4.33%±1.41 |
| Tax Agent BenchAgentic US corporate tax research | 25.17%±3.06 | 2.49%±1.02 |
| MedCodeMedical billing code support | 40.45%±2.12 | 31.33%±1.95 |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | N/A | 0.00%±0.00 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 11.11%±1.34 | 0.00%±0.00 |
| Vibe Code Bench v1.1Building web apps from scratch | 49.93%±4.77 | 3.68%±1.41 |
Benchmarks
Legal Research Bench *
DeepSeek V4
0.00%± 2.93
(84/84)Mercury 2.5
0.00%± 1.41
(84/84)Tax Agent Bench *
DeepSeek V4
0.00%± 3.06
(75/75)Mercury 2.5
0.00%± 1.02
(75/75)MedCode *
DeepSeek V4
0.00%± 2.12
(107/107)Mercury 2.5
0.00%± 1.95
(107/107)Terminal-Bench-Science
DeepSeek V4
N/AMercury 2.5
0.00%± 0.00
(42/42)Terminal-Bench 4.0
DeepSeek V4
0.00%± 1.34
(54/54)Mercury 2.5
0.00%± 0.00
(54/54)Vibe Code Bench v1.1 *
DeepSeek V4
0.00%± 4.77
(118/118)Mercury 2.5
0.00%± 1.41
(118/118)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-22
Comparison by Industry
Model performance on different sections of the economy.
Legal
51.70%
vs.43.69%
Finance
48.62%
vs.2.49%
Healthcare
57.80%
vs.43.21%
Math
16.00%
vs.3.00%
Science
N/A
vs.0.00%
Academic
88.32%
vs.N/A
Coding
46.20%
vs.6.05%
Cyber
N/A
vs.45.48%
Social Mobility
62.92%
vs.45.26%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
N/A
N/A
Cost / TestVals Index
N/A
vs.N/A
Input Cost/ 1M Tokens
$1.32
vs.$0.20
Input Cache Read/ 1M Tokens
$0.044
vs.$0.02
Output Cost/ 1M Tokens
$3.96
vs.$0.75
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | DeepSeek | Inception |
| Latency | N/A | N/A |
| Cost (In/Out) | $1.32 / $3.96 | $0.2 / $0.75 |
| Context Window | 1M | 260k |
| Max Output Token | 384k | 66k |
| Input Modality |