Compare Models
Select models (max 5)
Claude Haiku 5.5Mistral Large 4
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 54.31%±1.41 | 48.05%±1.11 |
| Legal Research BenchAgentic US legal research | 43.27%±3.44 | 31.73%±3.23 |
| Finance Agent (v2)Core financial analyst tasks | 54.14%±2.13 | 54.68%±0.58 |
| Tax Agent BenchAgentic US corporate tax research | 62.20%±3.22 | 63.29%±3.23 |
| MedCodeMedical billing code support | N/A | 40.69%±2.17 |
| Code MigrationRewriting programs in new languages | N/A | 30.56%±4.22 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 35.35%±2.20 | 22.73%±0.88 |
| Vibe Code Bench v1.1Building web apps from scratch | 90.44%±1.73 | 78.40%±3.55 |
Benchmarks
Vals Index *
Claude Haiku 5.5
0.00%± 1.41
(45/45)Mistral Large 4
0.00%± 1.11
(45/45)Legal Research Bench *
Claude Haiku 5.5
0.00%± 3.44
(75/75)Mistral Large 4
0.00%± 3.23
(75/75)Finance Agent (v2) *
Claude Haiku 5.5
0.00%± 2.13
(76/76)Mistral Large 4
0.00%± 0.58
(76/76)Tax Agent Bench *
Claude Haiku 5.5
0.00%± 3.22
(67/67)Mistral Large 4
0.00%± 3.23
(67/67)MedCode *
Claude Haiku 5.5
N/AMistral Large 4
0.00%± 2.17
(105/105)Code Migration *
Claude Haiku 5.5
N/AMistral Large 4
0.00%± 4.22
(74/74)Terminal-Bench 4.0
Claude Haiku 5.5
0.00%± 2.20
(45/45)Mistral Large 4
0.00%± 0.88
(45/45)Vibe Code Bench v1.1 *
Claude Haiku 5.5
0.00%± 1.73
(110/110)Mistral Large 4
0.00%± 3.55
(110/110)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-07
Comparison by Industry
Model performance on different sections of the economy.
Legal
22.26%
vs.23.78%
Finance
58.45%
vs.58.06%
Healthcare
N/A
vs.60.56%
Math
86.00%
vs.10.00%
Science
N/A
vs.39.60%
Coding
57.69%
vs.38.25%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$2.991
$13.78
Cost / TestVals Index
$2.991
vs.$13.78
Input Cost/ 1M Tokens
$0.10
vs.$1.36
Input Cache Write/ 1M Tokens
$0.125
vs.N/A
Input Cache Read/ 1M Tokens
$0.01
vs.$0.14
Output Cost/ 1M Tokens
$0.50
vs.$4.18
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Anthropic | Mistral AI |
| Latency | 2836.53s | 6350.41s |
| Cost (In/Out) | $0.1 / $0.5 | $1.36 / $4.18 |
| Context Window | 1M | 512k |
| Max Output Token | 128k | 256k |
| Input Modality |