Sign InOpen Brain
VercelEngineering PostOfficial Source

TypeSafe AI's Jev now available on AI Gateway

Jev gives agent loops typed decisions and probabilities without generating prose, offering a smaller control primitive for routing tools, retries, reviews, and guardrails.

Vercel · Sep 16, 2026
Open Source Open MarkdownOpen JSON
Source Summary

Jev accepts shared state and evaluates named questions in parallel, returning typed **Choice, Score, and Boolean** answers with probabilities. It is available through the experimental evaluate API in **AI SDK 7.0.105+**.

Practical Implication

Use it for narrow control-plane decisions in agent loops: tool or subagent selection, retry versus stop, risk scoring, and escalation to human review. Preserve question IDs and calibrate its confidence against labeled workflow examples.

Agent-Ready Context
Jev accepts shared state and evaluates named questions in parallel, returning typed **Choice, Score, and Boolean** answers with probabilities. It is available through the experimental evaluate API in **AI SDK 7.0.105+**.

Use it for narrow control-plane decisions in agent loops: tool or subagent selection, retry versus stop, risk scoring, and escalation to human review. Preserve question IDs and calibrate its confidence against labeled workflow examples.

The reported **193.6x speed** and **444.6x cost** advantages come from TypeSafe's own workflow evaluations, so they do not establish performance on your workload. The API is also experimental.
Connected Context · Feed7 Judgment

Jev narrows agent-model work to fast, typed control-plane judgments rather than open-ended execution. Its question IDs, probabilities, and parallel evaluation could make routing, stopping, and escalation observable and calibratable, but they complement rather than replace deterministic permissions, idempotency, validation, and recovery. Vendor-reported gains and an experimental API require local labeled evaluation.

Stop AI Agent Hallucinations: 5 Techniques + Production Patterns - Elizabeth Fuentes, AWSJev can supply typed risk or routing judgments, while Composio’s permissions, action records, verification, and preflight checks remain the enforcement layer around any resulting action.AI Agents Are Just Distributed Systems Now — Salman Munaf, TikTokA retry-versus-stop score may guide policy, but it cannot resolve unknown external outcomes; idempotency, reconciliation, bounded retries, and recovery are still required at tool boundaries.Your Agent Just Authorized What?! — Jay Mok & Ben Coumes, PaypalJev’s Boolean, Choice, and Score outputs could help classify escalation needs, but PayPal’s evidence shows consequential authorization must still be scoped, attributable, time-bound, and independently verifiable.
Context Map
agentcoding#harness-engineering#tool-use#agent-reliability
Uncertainty
The reported **193.6x speed** and **444.6x cost** advantages come from TypeSafe's own workflow evaluations, so they do not establish performance on your workload. The API is also experimental.