Tinker Serverless Inference
Beta pay-per-token hosted inference for Thinking Machines' open-weight Inkling and Inkling-Small models, priced separately from Tinker training; production use is by waitlist.
Find alternatives to Tinker Serverless Inference
What it does
- Model inference
Sources
- Pricing
-
Official documentation · 29 Oct 2026
"Serverless inference prices in US dollars per million tokens, as of October 19, 2026".
- Description
-
Official documentation · 29 Oct 2026
"Serverless inference is currently in beta and available for Inkling and Inkling-Small only. We do not recommend it for intensive production use until it is out of beta."
- What it does
-
Official documentation · 29 Oct 2026
"Serverless inference is currently in beta and available for Inkling and Inkling-Small only. We do not recommend it for intensive production use until it is out of beta." Table lists Tinker IDs and per-million-token prefill and sample prices.
- Sold within
-
Official documentation · 29 Oct 2026
A section of the Tinker docs and pricing page, not a separate site; its own price table, available only for Tinker's own Inkling models.
- Name
-
Official documentation · 29 Oct 2026
Docs heading "Serverless Inference (Beta)".
- URL
-
Official documentation · 29 Oct 2026
Own "Serverless inference" section with its own price table on the Tinker pricing page.
- Deployment
-
Official documentation · 29 Oct 2026
Hosted by Thinking Machines: "Serverless inference" under the Tinker cloud API.
Also from Thinking Machines Lab
-
Tinker API service
Training API for LoRA fine-tuning and sampling of open-weight models, including Thinking Machines' own Inkling family, with the training infrastructure run by Thinking Machines.
Open Thinking Machines Lab in the directory
Something wrong here? Send a correction — quote this product id: tinker-serverless-inference.