Sakhanda Wire
NVDA $153.59 -1.64% MSFT $507.00 -2.12% GOOGL $279.98 -1.32% META $619.94 -1.03% AMZN $256.62 -0.44%
← Back to the news

Jev turns agent decisions into typed probabilities

TypeSafe AI has released Jev, its first System One model, for making structured decisions inside AI-agent workflows. Unlike a conversational model, Jev does not generate text, write code or summarize documents. It receives unstructured text or JSON together with typed questions, then returns decisions, probabilities and confidence values.

The system offers three primitives: Choice selects one option and assigns probabilities to all options; Score evaluates a state against ordered rubric levels; and Noul estimates how likely a statement is to be true. Questions are processed in parallel in one request. TypeSafe says Jev is trained with Reinforcement Learning for Calibrated Decisions, intended to align confidence with accuracy. Choice supports up to 255 options.

The article examines 20 possible agentic applications, including model routing, tool-call approval, ticket triage, reranking, relevance checks and screening for prompt injection. TypeSafe reports that its workflow evaluations found Jev up to 193.6 times faster and 444.6 times cheaper, using GPT-6 Astra and Fable 5.1 as reference answers. The company says the results represent the higher end of gains seen in real-world workflows. Input costs $0.042 per million tokens, while output is free.

Written by Sakhanda Wire AI from reporting by MarkTechPost. Facts, quotes and full context are in the original article below.

Originally published by MarkTechPost on

Read the original on MarkTechPost ↗

The reporting and the image belong to MarkTechPost and are credited and linked above; the text on this page is Sakhanda Wire's own summary.

← Back to the news

More stories

All the latest news