Compare Models
Select models (max 5)
Claude Sonnet 4.5 (Nonthinking)Claude Haiku 5.5
 |  | |
|---|---|---|
| Vals IndexGDP-weighted benchmark | N/A | 54.31%±1.41 |
| Legal Research BenchAgentic US legal research | N/A | 43.27%±3.44 |
| Finance Agent (v2)Core financial analyst tasks | N/A | 54.14%±2.13 |
| Tax Agent BenchAgentic US corporate tax research | N/A | 62.20%±3.22 |
| MedCodeMedical billing code support | 40.57%±2.00 | N/A |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | N/A | 35.35%±2.20 |
| Vibe Code Bench v1.1Building web apps from scratch | N/A | 90.44%±1.73 |
Benchmarks
Vals Index *
Claude Sonnet 4.5 (Nonthinking)
N/AClaude Haiku 5.5
0.00%± 1.41
(45/45)Legal Research Bench *
Claude Sonnet 4.5 (Nonthinking)
N/AClaude Haiku 5.5
0.00%± 3.44
(75/75)Finance Agent (v2) *
Claude Sonnet 4.5 (Nonthinking)
N/AClaude Haiku 5.5
0.00%± 2.13
(76/76)Tax Agent Bench *
Claude Sonnet 4.5 (Nonthinking)
N/AClaude Haiku 5.5
0.00%± 3.22
(67/67)MedCode *
Claude Sonnet 4.5 (Nonthinking)
0.00%± 2.00
(105/105)Claude Haiku 5.5
N/ATerminal-Bench 4.0
Claude Sonnet 4.5 (Nonthinking)
N/AClaude Haiku 5.5
0.00%± 2.20
(45/45)Vibe Code Bench v1.1 *
Claude Sonnet 4.5 (Nonthinking)
N/AClaude Haiku 5.5
0.00%± 1.73
(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
N/A
vs.22.26%
Finance
N/A
vs.58.45%
Healthcare
62.54%
vs.N/A
Math
N/A
vs.86.00%
Education
32.88%
vs.N/A
Coding
N/A
vs.57.69%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
N/A
$2.991
Cost / TestVals Index
N/A
vs.$2.991
Input Cost/ 1M Tokens
$3.00
vs.$0.10
Input Cache Write/ 1M Tokens
$3.75
vs.$0.125
Input Cache Read/ 1M Tokens
$0.30
vs.$0.01
Output Cost/ 1M Tokens
$15.00
vs.$0.50
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Anthropic | Anthropic |
| Latency | N/A | 2836.53s |
| Cost (In/Out) | $3 / $15 | $0.1 / $0.5 |
| Context Window | 1M | 1M |
| Max Output Token | 64k | 128k |
| Input Modality |