$0.95 per 1M input tokens and $4.05 per 1M output tokens (cheapest host: DeepInfra). A typical benchmark task costs about $0.60 at xhigh effort.
Is Inkling free to use?›
There is no free API; the cheapest paid price is $0.95 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Inkling, and what is it best at?›
#76 of 336 in Chat & Knowledge, #56 of 134 in Coding, #86 of 135 in Agents & Tool Use, #73 of 197 in Writing & Language, #61 of 125 in Vision & Documents. Strongest at: Reasoning, Science, Long context, Knowledge.
Is Inkling better than Inkling Small?›
On ARC-AGI-1 · semi-private, Inkling scores 79.5 vs 84 — lower.
How can I use Inkling?›
Through the API of DeepInfra, Together.
What is the context window of Inkling?›
256K tokens.
When was Inkling released? Is there a newer version?›
Released 2026-07-15. It is the current version in its line.
Is Inkling open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Inkling locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Inkling read images?›
It accepts text and images, audio as input.
What is Inkling?›
Inkling is an open-weights language model from Thinking Machines, released Jul 2026.