Sakhanda Wire
NVDA $153.64 -1.60% MSFT $506.98 -2.13% GOOGL $279.95 -1.33% META $620.09 -1.01% AMZN $256.57 -0.46%
← Back to the news

TypeSafe AI launches Jev for typed, probabilistic software decisions

TypeSafe AI has released Jev, a transformer-based System One model designed to return typed decisions rather than generated text. Applications send a state and typed questions, then receive answers with probabilities that software can use for branching. Jev is available as a hosted API in early access through a waitlist; TypeSafe has not disclosed its weights, parameter count, or a self-hosting option.

The API uses one endpoint and supports three question types: Choice selects from up to 255 options, Score rates an item against ordered levels, and Noul returns a probability from 0 to 1 for a statement. Questions run in parallel against the same state. Each answer includes a confidence value derived from its probability distribution. TypeSafe recommends acting automatically on high-confidence results, reviewing intermediate ones, and routing low-confidence cases to people, with thresholds tied to the cost of errors.

Input costs $0.042 per 1M tokens, while output tokens are free. TypeSafe reports response times of 70ms to 500ms and vendor-run evaluations claiming major speed and cost advantages over selected models. The company notes that these tests used its own workflows and reference answers, and that “zero hallucinations” refers only to schema matching; answers can still be wrong.

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