Jev API Pro is an independent workspace and REST API for TypeSafe's Jev model, accessed through OpenRouter (not affiliated with TypeSafe). It turns a piece of text into one decision from a list of labels you define, so your application can route it without parsing a chat response.
How it works
- Send POST /api/v1/decisions with the text, short instructions and 2-12 allowed labels.
- Get back one label, confidence and per-label probabilities when the provider supplies them, and a needsReview flag based on your threshold (default 0.85).
- Or ask up to four typed questions about the same text in one request: choice, true/false, or an ordered score.
- Idempotency keys make retries safe; a model failure returns an explicit error, never a default label.
Use cases Support ticket routing, intent classification, lead qualification, feedback tagging, and any workflow where low-confidence results should go to a human.
Workspace Web playground, batch CSV classification (up to 1,000 rows), decision history and API keys. Decision history stores the result and a request hash, not the submitted message.
Pricing Try it free: 3 anonymous attempts per day, 20 signup credits and 2 daily check-in credits. Prepaid plans (30-day, 90-day or annual) add API access and larger credit volumes.
