Compare Models
Select models (max 7)
Claude Sonnet 5.5GPT-6 Astra
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 70.85%±0.93 | 65.02%±1.10 |
| Legal Research BenchAgentic US legal research | 48.08%±3.47 | 39.42%±3.40 |
| Finance Agent v2.1Core financial analyst tasks | 67.28%±2.02 | 60.43%±2.12 |
| Tax Agent BenchAgentic US corporate tax research | 41.96%±3.56 | 20.72%±2.92 |
| MedCodeMedical billing code support | 52.92%±2.12 | 48.49%±2.13 |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 47.14%±6.01 | 62.86%±5.82 |
| Code Migration v1.1Rewriting programs in new languages | 86.88%±1.52 | 80.58%±2.34 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 64.14%±1.01 | 59.60%±4.40 |
| Vibe Code Bench v1.1Building web apps from scratch | 92.39%±1.26 | 89.59%±2.17 |
Benchmarks
Vals Index *
Claude Sonnet 5.5
0.00%± 0.93
(37/37)GPT-6 Astra
0.00%± 1.10
(37/37)Legal Research Bench *
Claude Sonnet 5.5
0.00%± 3.47
(94/94)GPT-6 Astra
0.00%± 3.40
(94/94)Finance Agent v2.1 *
Claude Sonnet 5.5
0.00%± 2.02
(40/40)GPT-6 Astra
0.00%± 2.12
(40/40)Tax Agent Bench *
Claude Sonnet 5.5
0.00%± 3.56
(85/85)GPT-6 Astra
0.00%± 2.92
(85/85)MedCode *
Claude Sonnet 5.5
0.00%± 2.12
(107/107)GPT-6 Astra
0.00%± 2.13
(107/107)Terminal-Bench-Science
Claude Sonnet 5.5
0.00%± 6.01
(45/45)GPT-6 Astra
0.00%± 5.82
(45/45)Code Migration v1.1 *
Claude Sonnet 5.5
0.00%± 1.52
(30/30)GPT-6 Astra
0.00%± 2.34
(30/30)Terminal-Bench 4.0
Claude Sonnet 5.5
0.00%± 1.01
(64/64)GPT-6 Astra
0.00%± 4.40
(64/64)Vibe Code Bench v1.1 *
Claude Sonnet 5.5
0.00%± 1.26
(128/128)GPT-6 Astra
0.00%± 2.17
(128/128)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, legal, and tax
•2026-11-11
Comparison by Industry
Model performance on different sections of the economy.
Legal
42.49%
vs.38.07%
Finance
71.28%
vs.51.28%
Healthcare
72.01%
vs.68.20%
Math
100.00%
vs.99.00%
Science
59.12%
vs.54.65%
Education
88.03%
vs.N/A
Coding
65.22%
vs.58.04%
Cyber
44.87%
vs.48.97%
Social Mobility
71.06%
vs.N/A
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$16.18
$21.28
Cost / TestVals Index
$16.18
vs.$21.28
Input Cost/ 1M Tokens
$2.00
vs.$10.00
Input Cache Write/ 1M Tokens
$2.50
vs.$12.50
Input Cache Read/ 1M Tokens
$0.10
vs.$1.00
Output Cost/ 1M Tokens
$10.00
vs.$50.00
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Anthropic | OpenAI |
| Latency | 5158.77s | 2018.88s |
| Cost (In/Out) | $2 / $10 | $10 / $50 |
| Context Window | 1M | 1M |
| Max Output Token | 128k | 128k |
| Input Modality |