ModelBeat
Models/Inkling Small
All models
T

Thinking Machines: Inkling Small

thinkingmachines/inkling-small

Inkling Small is Thinking Machines’s AI model, released Jul 30, 2026. It costs $0.58/M input and $1.44/M output tokens, and handles a 524K-token context window. Its strongest use case is coding, where it ranks in the 71st percentile of the models Model Beat tracks.

Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total.

Context
524K
Input
$0.58 / 1M
Output
$1.44 / 1M
Released
Jul 30, 2026
Modalities
textimageaudiotext
Strongest atCoding top 29%Reasoning & Knowledge top 31%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Inkling Small lands among all models tracked on Model Beat.

70.0
Intelligence Index
Artificial Analysis
70th percentile of tracked models
71.0
Coding Index
Artificial Analysis
71st percentile of tracked models

Reasoning

2 evals
GPQA Diamond89.5%

Graduate-level scientific reasoning

Humanity's Last Exam31.6%

Frontier of human expert knowledge

Coding

1 evals
SciCode48.7%

Scientific research coding

In the news

1

Changelog

1
  • Inkling Small (Thinking Machines) added to the tracker: $0.58/$1.44 per 1M, 524K context.tracked · Jul 31, 2026

Compare Inkling Small with

Frequently asked questions

What is Inkling Small?

Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total.

Who created Inkling Small?

Inkling Small is an AI model developed by Thinking Machines, released Jul 30, 2026.

How much does Inkling Small cost?

Inkling Small is served from $0.58 per million input tokens and $1.44 per million output tokens (cheapest credible provider on OpenRouter: DeepInfra).

What is Inkling Small's context window?

Inkling Small supports a context window of up to 524K tokens.

How does Inkling Small perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Inkling Small scores 89.5% on GPQA Diamond, 48.7% on SciCode, 31.6% on Humanity's Last Exam.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Aug 2, 2026.