Compare Models
Select models (max 5)
GPT 5.4 MiniInkling
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 33.17%±1.28 | 28.67%±1.10 |
| Legal Research BenchAgentic US legal research | 12.50%±2.30 | 28.36%±3.13 |
| Finance Agent (v2)Core financial analyst tasks | 45.36%±0.45 | 46.60%±0.93 |
| Tax Agent BenchAgentic US corporate tax research | 36.59%±3.10 | 43.52%±2.95 |
| MedCodeMedical billing code support | N/A | 41.19%±2.23 |
| Code MigrationRewriting programs in new languages | 12.94%±3.64 | 11.79%±3.40 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 2.52%±1.01 | 0.51%±0.51 |
| Vibe Code Bench v1.1Building web apps from scratch | 47.97%±5.61 | 19.21%±3.66 |
Benchmarks
Vals Index *
GPT 5.4 Mini
0.00%± 1.28
(43/43)Inkling
0.00%± 1.10
(43/43)Legal Research Bench *
GPT 5.4 Mini
0.00%± 2.30
(73/73)Inkling
0.00%± 3.13
(73/73)Finance Agent (v2) *
GPT 5.4 Mini
0.00%± 0.45
(74/74)Inkling
0.00%± 0.93
(74/74)Tax Agent Bench *
GPT 5.4 Mini
0.00%± 3.10
(65/65)Inkling
0.00%± 2.95
(65/65)MedCode *
GPT 5.4 Mini
N/AInkling
0.00%± 2.23
(104/104)Code Migration *
GPT 5.4 Mini
0.00%± 3.64
(73/73)Inkling
0.00%± 3.40
(73/73)Terminal-Bench 4.0
GPT 5.4 Mini
0.00%± 1.01
(43/43)Inkling
0.00%± 0.51
(43/43)Vibe Code Bench v1.1 *
GPT 5.4 Mini
0.00%± 5.61
(108/108)Inkling
0.00%± 3.66
(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
6.25%
vs.37.80%
Finance
52.42%
vs.53.55%
Healthcare
N/A
vs.63.30%
Math
N/A
vs.0.00%
Academic
82.29%
vs.83.97%
Education
50.81%
vs.36.55%
Coding
36.32%
vs.27.00%
Cyber
N/A
vs.66.79%
Social Mobility
N/A
vs.58.46%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$1.452
$1.464
Cost / TestVals Index
$1.452
vs.$1.464
Input Cost/ 1M Tokens
$0.75
vs.$1.00
Input Cache Read/ 1M Tokens
$0.075
vs.$0.17
Output Cost/ 1M Tokens
$4.50
vs.$4.05
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | OpenAI | Thinking Machines |
| Latency | 1788.54s | 1205.92s |
| Cost (In/Out) | $0.75 / $4.5 | $1 / $4.05 |
| Context Window | 400k | 256k |
| Max Output Token | 128k | 256k |
| Input Modality |