Compare Models
Select models (max 5)
Claude Opus 5GPT-6 Astra
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 63.67%±0.96 | 63.13%±1.19 |
| Legal Research BenchAgentic US legal research | 55.29%±3.46 | 39.42%±3.40 |
| Finance Agent (v2)Core financial analyst tasks | 58.63%±0.08 | 53.54%±2.08 |
| Tax Agent BenchAgentic US corporate tax research | 75.06%±2.96 | 63.34%±3.15 |
| MedCodeMedical billing code support | 63.57%±1.99 | 48.49%±2.13 |
| Terminal-Bench ScienceExpert-authored scientific research workflows | 27.14%±5.35 | 62.86%±5.82 |
| Code MigrationRewriting programs in new languages | 57.47%±4.37 | 67.74%±4.22 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 53.53%±1.34 | 59.60%±4.40 |
| Vibe Code Bench v1.1Building web apps from scratch | 88.40%±3.00 | 89.59%±2.17 |
Benchmarks
Vals Index *
Claude Opus 5
0.00%± 0.96
(43/43)GPT-6 Astra
0.00%± 1.19
(43/43)Legal Research Bench *
Claude Opus 5
0.00%± 3.46
(73/73)GPT-6 Astra
0.00%± 3.40
(73/73)Finance Agent (v2) *
Claude Opus 5
0.00%± 0.08
(74/74)GPT-6 Astra
0.00%± 2.08
(74/74)Tax Agent Bench *
Claude Opus 5
0.00%± 2.96
(65/65)GPT-6 Astra
0.00%± 3.15
(65/65)MedCode *
Claude Opus 5
0.00%± 1.99
(104/104)GPT-6 Astra
0.00%± 2.13
(104/104)Terminal-Bench Science
Claude Opus 5
0.00%± 5.35
(38/38)GPT-6 Astra
0.00%± 5.82
(38/38)Code Migration *
Claude Opus 5
0.00%± 4.37
(73/73)GPT-6 Astra
0.00%± 4.22
(73/73)Terminal-Bench 4.0
Claude Opus 5
0.00%± 1.34
(43/43)GPT-6 Astra
0.00%± 4.40
(43/43)Vibe Code Bench v1.1 *
Claude Opus 5
0.00%± 3.00
(108/108)GPT-6 Astra
0.00%± 2.17
(108/108)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-02
Comparison by Industry
Model performance on different sections of the economy.
Legal
49.64%
vs.22.42%
Finance
70.89%
vs.62.86%
Healthcare
77.28%
vs.68.20%
Math
99.00%
vs.99.00%
Science
47.93%
vs.65.09%
Academic
91.64%
vs.N/A
Education
49.43%
vs.46.37%
Coding
62.42%
vs.58.35%
Cyber
38.79%
vs.48.97%
Social Mobility
76.93%
vs.N/A
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$19.31
$18.46
Cost / TestVals Index
$19.31
vs.$18.46
Input Cost/ 1M Tokens
$5.00
vs.$10.00
Input Cache Write/ 1M Tokens
$6.25
vs.$12.50
Input Cache Read/ 1M Tokens
$0.50
vs.$1.00
Output Cost/ 1M Tokens
$25.00
vs.$50.00
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Anthropic | OpenAI |
| Latency | 3461.38s | 1667.74s |
| Cost (In/Out) | $5 / $25 | $10 / $50 |
| Context Window | 1M | 1M |
| Max Output Token | 128k | 128k |
| Input Modality |