# TypeSafe AI's Jev 1.14 decision model is now available on AI Gateway

published: 2026-10-15T17:06:13.104Z
topic: infrastructure
source: https://vercel.com/changelog/typesafe-ais-jev-1-14-is-now-available-on-ai-gateway
url: https://aiagentnews.io/stories/typesafe-ai-s-jev-1-14-decision-model-is-now-available-on-ai-gateway

Jev 1.14, from TypeSafe AI, is now available on Vercel's AI Gateway. The model returns typed Choice, Score and Boolean answers with probabilities for a set of questions in a single parallel request.

Jev 1.14 from TypeSafe AI is now available on AI Gateway. Jev is a decision model: developers pass it state and a list of questions, and it returns typed answers with probabilities instead of free-form text. It answers every question in one request, in parallel, and is aimed at classification, routing, scoring and automated verification.

This version accepts images as part of the state, so a file part can sit alongside text. It also adds an experimental long-context mode for documents, logs or conversation histories that are too long for a standard request, enabled with providerOptions.typesafe.longContext set to true. The model name is typesafe-ai/jev-1.14.0, and the examples call it through experimental_decide in the AI SDK.

Jev supports Zero Data Retention and No Training, which can be turned on for each request. Decision calls appear in logs and custom reporting, count toward budgets, and accept other Gateway provider options through the providerOptions.gateway object.

## Why it matters if you build agents

Agent builders can get several typed answers from one call for routing, scoring or verification, and can now pass images as part of that check. Long-context mode is marked experimental, so test it on your own logs or documents before relying on it. Since decision calls count toward budgets, include them when tracking model spend.