Jev + OpenRouter
Using Jev on OpenRouter
OpenRouter hosts Jev alongside hundreds of language models, but Jev doesn't use the familiar chat completions endpoint. Because it returns decisions rather than text, OpenRouter serves it through a dedicated Decisions API.
Get Early Access to ClassifierHub: 2× credits in your first paid month.
Getting access
Create an OpenRouter account, add credits and generate an API key. Jev is billed to your OpenRouter balance like other models. Because it doesn't produce text, it may not appear in default model lists that filter on text output.
- Model id: typesafe/jev-1.13
- Latest alias: ~typesafe/jev-latest
- Endpoint: POST https://openrouter.ai/api/alpha/decisions
- Context window: about 32,000 tokens
How it differs from chat completions
There are no messages, system prompts or temperature settings. You send a state and typed questions. There is no streaming, because the answer is a small structured object. Output tokens are free, so cost is driven by input size.
Going beyond the raw endpoint
The raw endpoint is ideal if you want full control. If you'd rather not build API key management, usage metering, templates, batch jobs and agent integrations yourself, ClassifierHub provides those on top of Jev with a stable, versioned API.
Frequently asked questions
Related guides
Last updated 2026-09-25. ClassifierHub is an independent product built on top of the Jev decision model, accessed through OpenRouter. It is not affiliated with or endorsed by TypeSafe or OpenRouter.