Compare Models
Select models (max 7)
GPT-6 LunaStep 5 Preview
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 52.32%±1.12 | N/A |
| Legal Research BenchAgentic US legal research | 30.29%±3.19 | 40.38%±3.41 |
| Finance Agent v2.1Core financial analyst tasks | 56.84%±2.12 | 58.90%±2.05 |
| Tax Agent BenchAgentic US corporate tax research | 19.95%±2.87 | 27.27%±3.11 |
| MedCodeMedical billing code support | 44.69%±2.30 | N/A |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 4.29%±2.44 | 2.86%±2.01 |
| Code Migration v1.1Rewriting programs in new languages | 46.67%±3.61 | 47.00%±3.43 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 13.64%±1.51 | 31.82%±2.62 |
| Vibe Code Bench v1.1Building web apps from scratch | 81.65%±3.38 | 70.40%±4.27 |
Benchmarks
Vals Index *
GPT-6 Luna
0.00%± 1.12
(26/26)Step 5 Preview
N/ALegal Research Bench *
GPT-6 Luna
0.00%± 3.19
(84/84)Step 5 Preview
0.00%± 3.41
(84/84)Finance Agent v2.1 *
GPT-6 Luna
0.00%± 2.12
(30/30)Step 5 Preview
0.00%± 2.05
(30/30)Tax Agent Bench *
GPT-6 Luna
0.00%± 2.87
(75/75)Step 5 Preview
0.00%± 3.11
(75/75)MedCode *
GPT-6 Luna
0.00%± 2.30
(107/107)Step 5 Preview
N/ATerminal-Bench-Science
GPT-6 Luna
0.00%± 2.44
(42/42)Step 5 Preview
0.00%± 2.01
(42/42)Code Migration v1.1 *
GPT-6 Luna
0.00%± 3.61
(28/28)Step 5 Preview
0.00%± 3.43
(28/28)Terminal-Bench 4.0
GPT-6 Luna
0.00%± 1.51
(54/54)Step 5 Preview
0.00%± 2.62
(54/54)Vibe Code Bench v1.1 *
GPT-6 Luna
0.00%± 3.38
(118/118)Step 5 Preview
0.00%± 4.27
(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
17.23%
vs.40.38%
Finance
61.48%
vs.51.91%
Healthcare
64.20%
vs.N/A
Math
64.00%
vs.42.00%
Science
28.38%
vs.29.58%
Education
48.09%
vs.N/A
Coding
39.25%
vs.45.62%
Cyber
39.46%
vs.0.38%
Social Mobility
57.65%
vs.58.12%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$0.45
N/A
Cost / TestVals Index
$0.45
vs.N/A
Input Cost/ 1M Tokens
$0.10
vs.$1.00
Input Cache Write/ 1M Tokens
$0.125
vs.N/A
Input Cache Read/ 1M Tokens
$0.01
vs.$0.05
Output Cost/ 1M Tokens
$0.50
vs.$2.70
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | OpenAI | Stepfun |
| Latency | 1943.19s | N/A |
| Cost (In/Out) | $0.1 / $0.5 | $1 / $2.7 |
| Context Window | 1M | 1M |
| Max Output Token | 128k | 1M |
| Input Modality |