Compare Models

Select models (max 7)
Inkling 2 MiniGemini 4 Neon
Benchmarks

Vals Index *

Inkling 2 Mini
0.00%± 1.11
(43/43)
Gemini 4 Neon
0.00%± 0.94
(43/43)

Legal Research Bench *

Inkling 2 Mini
0.00%± 2.66
(100/100)
Gemini 4 Neon
0.00%± 3.45
(100/100)

Finance Agent v2.1 *

Inkling 2 Mini
0.00%± 2.13
(46/46)
Gemini 4 Neon
0.00%± 1.94
(46/46)

Tax Agent Bench *

Inkling 2 Mini
0.00%± 2.20
(91/91)
Gemini 4 Neon
0.00%± 3.59
(91/91)

MedCode *

Inkling 2 Mini
0.00%± 1.90
(109/109)
Gemini 4 Neon
N/A

Terminal-Bench-Science

Inkling 2 Mini
0.00%± 0.00
(49/49)
Gemini 4 Neon
0.00%± 5.82
(49/49)

Code Migration v1.1 *

Inkling 2 Mini
0.00%± 3.87
(31/31)
Gemini 4 Neon
N/A

Terminal-Bench 4.0

Inkling 2 Mini
0.00%± 0.51
(70/70)
Gemini 4 Neon
0.00%± 2.02
(70/70)

Vibe Code Bench v1.1 *

Inkling 2 Mini
0.00%± 3.76
(134/134)
Gemini 4 Neon
0.00%± 1.17
(134/134)

Overall performance

Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, legal, and tax

2026-11-19

Comparison by Industry

Model performance on different sections of the economy.

Legal
11.39%
37.92%
Finance
45.56%
66.64%
Healthcare
56.04%
N/A
Math
18.00%
N/A
Science
23.58%
62.63%
Education
85.89%
N/A
Coding
28.70%
65.87%
Cyber
37.29%
N/A

Cost Analysis

Per task and per token model pricing.

Vals Index · USD
$3.197
$7.405
Cost / TestVals Index
$3.197
$7.405
Input Cost/ 1M Tokens
$0.15
$3.00
Input Cache Read/ 1M Tokens
$0.03
$0.15
Output Cost/ 1M Tokens
$0.50
$15.00

Model Metadata

Basic information about each model.

Model provider
Thinking MachinesThinking Machines
GoogleGoogle
Latency2835.36s1884.68s
Cost (In/Out)$0.15 / $0.5$3 / $15
Context Window1M1M
Max Output Token1M1M
Input Modality