Compare Models
Select models (max 7)
Gemini 4 NeonGPT-6.2 Astra
| Vals IndexGDP-weighted benchmark | 70.87%±0.94 | 67.66%±0.94 |
| Legal Research BenchAgentic US legal research | 56.25%±3.45 | 42.79%±3.44 |
| Finance Agent v2.1Core financial analyst tasks | 68.56%±1.94 | 60.75%±2.12 |
| Tax Agent BenchAgentic US corporate tax research | 54.89%±3.59 | 21.84%±2.79 |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 62.86%±5.82 | 71.43%±5.44 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 61.62%±2.02 | 63.13%±1.34 |
| Vibe Code Bench v1.1Building web apps from scratch | 93.70%±1.17 | 89.90%±2.04 |
Benchmarks
Vals Index *
Gemini 4 Neon
0.00%± 0.94
(42/42)GPT-6.2 Astra
0.00%± 0.94
(42/42)Legal Research Bench *
Gemini 4 Neon
0.00%± 3.45
(99/99)GPT-6.2 Astra
0.00%± 3.44
(99/99)Finance Agent v2.1 *
Gemini 4 Neon
0.00%± 1.94
(45/45)GPT-6.2 Astra
0.00%± 2.12
(45/45)Tax Agent Bench *
Gemini 4 Neon
0.00%± 3.59
(90/90)GPT-6.2 Astra
0.00%± 2.79
(90/90)Terminal-Bench-Science
Gemini 4 Neon
0.00%± 5.82
(48/48)GPT-6.2 Astra
0.00%± 5.44
(48/48)Terminal-Bench 4.0
Gemini 4 Neon
0.00%± 2.02
(69/69)GPT-6.2 Astra
0.00%± 1.34
(69/69)Vibe Code Bench v1.1 *
Gemini 4 Neon
0.00%± 1.17
(133/133)GPT-6.2 Astra
0.00%± 2.04
(133/133)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, legal, and tax
Comparison by Industry
Model performance on different sections of the economy.
Legal
37.92%
vs.29.52%
Finance
66.64%
vs.52.87%
Math
N/A
vs.87.50%
Science
62.63%
vs.68.44%
Coding
77.66%
vs.69.39%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$7.405
$10.68
Cost / TestVals Index
$7.405
vs.$10.68
Input Cost/ 1M Tokens
$3.00
vs.$8.00
Input Cache Write/ 1M Tokens
N/A
vs.$10.00
Input Cache Read/ 1M Tokens
$0.15
vs.$0.40
Output Cost/ 1M Tokens
$15.00
vs.$40.00
Model Metadata
Basic information about each model.
| Model provider | ||
| Latency | 1884.68s | 3136.31s |
| Cost (In/Out) | $3 / $15 | $8 / $40 |
| Context Window | 1M | 1M |
| Max Output Token | 1M | 128k |
| Input Modality |