Inkling 2 Small
by Thinking Machines · 594B total / 17B active parameters
Today's price · per 1M tokens
Input
$0.300
per 1M tokens
Output
$1.00
per 1M tokens
Blended
$0.475
blended $/1M — 3:1 weighted input:output
Cached input
$0.030
10% of input — prompt caching
How this price is scoped: Tinker Serverless Inference (Beta) list price for the 256K-context nvfp4 sampling endpoint. The 64K base id on Tinker's training table prints the same inference rates.
Price receipt
We read Thinking Machines's own pricing page on and found Inkling 2 Small listed with a price on the same row — that read is where the number above comes from, recorded as a new model. We keep the page text we read; its content hash is 9daf160046.
Overview
Thinking Machines' second model generation: a 594B-total / 17B-active Mixture-of-Experts with image input and variable thinking effort, trained with a large unified RL run on long-horizon agentic coding. Unlike earlier Inkling models it does not take audio. It launched on Tinker with closed weights; the lab says it intends to release them under Apache 2.0 after external safety testing. Sold through Tinker serverless inference at $0.30/$1.00 per 1M tokens, with cached input at $0.03.
Capabilities
struck through = not supportedSpecifications
- Provider
- Thinking Machines
- Context window
- 262K tokens
- Modality
- text, image
- Parameters
- 594B total / 17B active
- Open source
- No — proprietary
- Released
- Status
- Current
- Last updated
- Tags
Availability verified: Not re-verified. We have no recent evidence that Thinking Machines still lists this model — treat availability as unconfirmed.