Compare Models
Select models (max 7)
Mistral Large 4DeepSeek V4.1 Flash
 |  |  |
|---|---|---|
| Legal Research BenchAgentic US legal research | 31.73%±3.23 | 41.35%±3.42 |
| Finance Agent v2.1Core financial analyst tasks | 61.38%±2.02 | 65.43%±2.02 |
| Tax Agent BenchAgentic US corporate tax research | 30.39%±3.31 | 26.63%±3.14 |
| MedCodeMedical billing code support | 40.69%±2.17 | 41.17%±2.04 |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | N/A | 0.00%±0.00 |
| Code Migration v1.1Rewriting programs in new languages | N/A | 58.94%±2.66 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 27.78%±1.82 | 19.70%±1.75 |
| Vibe Code Bench v1.1Building web apps from scratch | 78.40%±3.55 | 84.74%±2.89 |
Benchmarks
Legal Research Bench *
Mistral Large 4
0.00%± 3.23
(89/89)DeepSeek V4.1 Flash
0.00%± 3.42
(89/89)Finance Agent v2.1 *
Mistral Large 4
0.00%± 2.02
(35/35)DeepSeek V4.1 Flash
0.00%± 2.02
(35/35)Tax Agent Bench *
Mistral Large 4
0.00%± 3.31
(80/80)DeepSeek V4.1 Flash
0.00%± 3.14
(80/80)MedCode *
Mistral Large 4
0.00%± 2.17
(107/107)DeepSeek V4.1 Flash
0.00%± 2.04
(107/107)Terminal-Bench-Science
Mistral Large 4
N/ADeepSeek V4.1 Flash
0.00%± 0.00
(43/43)Code Migration v1.1 *
Mistral Large 4
N/ADeepSeek V4.1 Flash
0.00%± 2.66
(29/29)Terminal-Bench 4.0
Mistral Large 4
0.00%± 1.82
(59/59)DeepSeek V4.1 Flash
0.00%± 1.75
(59/59)Vibe Code Bench v1.1 *
Mistral Large 4
0.00% ± 3.55
(123/123)DeepSeek V4.1 Flash
0.00%± 2.89
(123/123)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-29
Comparison by Industry
Model performance on different sections of the economy.
Legal
37.65%
vs.63.42%
Finance
53.58%
vs.53.27%
Healthcare
60.56%
vs.63.34%
Math
10.00%
vs.54.00%
Science
39.60%
vs.29.65%
Education
N/A
vs.85.38%
Coding
39.77%
vs.42.10%
Cyber
N/A
vs.37.23%
Social Mobility
N/A
vs.64.28%
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
$0.70
vs.$0.30
Input Cache Read/ 1M Tokens
$0.07
vs.$0.006
Output Cost/ 1M Tokens
$2.10
vs.$1.20
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Mistral AI | DeepSeek |
| Latency | N/A | N/A |
| Cost (In/Out) | $0.7 / $2.1 | $0.3 / $1.2 |
| Context Window | 512k | 1M |
| Max Output Token | 256k | 384k |
| Input Modality |