Compare Models
Select models (max 7)
Qwen 3.8 MaxMistral Large 4
 |  |  |
|---|---|---|
| Legal Research BenchAgentic US legal research | 47.60%±3.47 | 31.73%±3.23 |
| Finance Agent v2.1Core financial analyst tasks | N/A | 61.38%±2.02 |
| Tax Agent BenchAgentic US corporate tax research | 32.09%±3.37 | 30.39%±3.31 |
| MedCodeMedical billing code support | 40.67%±2.03 | 40.69%±2.17 |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 12.86%±4.03 | N/A |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 34.34%±3.94 | 27.78%±1.82 |
| Vibe Code Bench v1.1Building web apps from scratch | 64.70%±5.36 | 78.40%±3.55 |
Benchmarks
Legal Research Bench *
Qwen 3.8 Max
0.00%± 3.47
(84/84)Mistral Large 4
0.00%± 3.23
(84/84)Finance Agent v2.1 *
Qwen 3.8 Max
N/AMistral Large 4
0.00%± 2.02
(30/30)Tax Agent Bench *
Qwen 3.8 Max
0.00%± 3.37
(75/75)Mistral Large 4
0.00%± 3.31
(75/75)MedCode *
Qwen 3.8 Max
0.00%± 2.03
(107/107)Mistral Large 4
0.00%± 2.17
(107/107)Terminal-Bench-Science
Qwen 3.8 Max
0.00%± 4.03
(42/42)Mistral Large 4
N/ATerminal-Bench 4.0
Qwen 3.8 Max
0.00%± 3.94
(54/54)Mistral Large 4
0.00%± 1.82
(54/54)Vibe Code Bench v1.1 *
Qwen 3.8 Max
0.00%± 5.36
(118/118)Mistral Large 4
0.00%± 3.55
(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
65.60%
vs.24.62%
Finance
59.41%
vs.53.58%
Healthcare
62.81%
vs.60.56%
Math
58.00%
vs.10.00%
Science
18.37%
vs.39.60%
Academic
90.11%
vs.N/A
Education
51.25%
vs.N/A
Coding
48.75%
vs.39.77%
Cyber
28.57%
vs.N/A
Social Mobility
67.12%
vs.N/A
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
$2.00
vs.$1.36
Input Cache Write/ 1M Tokens
$2.50
vs.N/A
Input Cache Read/ 1M Tokens
$0.25
vs.$0.14
Output Cost/ 1M Tokens
$6.00
vs.$4.18
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Alibaba | Mistral AI |
| Latency | N/A | N/A |
| Cost (In/Out) | $2 / $6 | $1.36 / $4.18 |
| Context Window | 1M | 512k |
| Max Output Token | 128k | 256k |
| Input Modality |