Inkling Small
Thinking Machines Lab馃嚭馃嚫
thinkingmachines/inkling-smallInkling-Small is a compact open-weights sibling of Thinking Machines Lab's Inkling model, at 276B total parameters (12B active) versus Inkling's 975B. It reasons natively over text, images, and audio with adjustable thinking effort, achieving performance close to the full Inkling model at a fraction of its size.
Cost rate
0.3x
Context
1M
Released
Jul 30, 2026
Input
TextImageAudio
Output
Text
Support
ReasoningTool callingStructured outputs
Best at
The categories where this model ranks highest.
Image understanding
#73 路 top 48%
Mathematical
#109 路 top 29%
Software & IT Services
#119 路 top 30%
Business, Management, & Finance
#138 路 top 35%
Performance
Median latency and throughput measured across recent requests.
Throughput184 tok/s
Latency