All models
View catalogue
TypeSafe AI
Verification
Jev
A fast decision model that returns typed choices instead of generated prose.
jev-latestAt a glance
- Published input cap
- —
- Published output cap
- —
- Base input / output per 1M
- Managed price not published
- Platform fee
- —
Rates are live billing reference prices, before the listed platform fee.Availability and features depend on your project and endpoint.
Overview
About Jev
A fast decision model that returns typed choices instead of generated prose. Use the live /v1/models endpoint for the capabilities available to your project. A published price does not by itself guarantee a usable provider route.
Model fit
Best for
- 01Automated policy enforcement, risk scoring, and real-time fraud decisions
- 02Instant intent classification and multi-agent workflow decision gates
- 03System One runtime evaluations verifying whether upstream LLM outputs meet safety specs
Integration
Use through System One
This model returns typed decisions rather than generated prose. Connect it through Gruvo's System One integration.
Use Jev for bounded choices, scoring, classification, routing, and verification where your application—not the model—defines the allowed outcomes.
Integration guideKeep comparing