Inkling-Small
A multimodal 276B-parameter open-weight MoE model with 12B active parameters.
Inkling-Small is described as a 276-billion-parameter open-weight mixture-of-experts model with 12 billion active parameters, multimodal inputs, and a 1-million-token context window.
Pricing
Official site · Oct 31, 2026, 1:08 PM
Input / 1M
$0.30
Output / 1M
$1.20
Cached input / 1M
$0.06
Serverless inference beta, 256K context; effective Oct. 19, 2026. Training/sampling rates are separate.
Official Tinker pricing page lists Inkling-Small serverless inference beta at USD per 1M tokens: $0.30 input, $0.06 cached input, and $1.20 output (256K context). Also lists separate training/sampling prices; normalized values recorded here are serverless inference rates. Model card identifies it as multimodal.
Model Intelligence
Context window
1,000,000 tokens
Arena ranking
13
Benchmarkable
Yes
Model level
release
Intelligence Index
12.9
GPQA
0.9
HLE
0.33
SciCode
0.5
LCR
0.76
Recent stories
0 linked stories
No linked stories yet.