Skip to main content
OneHot is Sqwish Labs’ AI decision engine pre-trained to produce calibrated, low-latency responses to questions in a variety of domains. Use the kettei model family directly, or fine-tune your own family to best suit your use case.

​
Three decision types

Mix types in one request. Results are keyed by your decision IDs. Add an abstention policy when uncertain cases should go to review.
Probabilities describe the alternatives you declared. They are not a guarantee that an answer is correct. Test your wording and action thresholds on representative cases before using them to automate work.

​
Try it

Access the free playground and experiment with the provided examples, or create your own.

Access the playground

Open the playground and experiment with our decision models.
For a more flexible experience, sign up for an account and use the OneHot API.

​
Where to go from here

The quickstart guide shows you how to use the OneHot API in your application. You can use the HTTP API directly, or try out our Python or TypeScript SDKs. You can also directly connect your coding agent to power your agents to make decisions through the OneHot MCP server and Claude Code tool-call hook. Your data is not used to train the models offered to anyone else. A model fine-tuned on your data can be called only by your account. For more information, see data residency, the privacy policy and the terms of service.