A word-level generation experiment that asks Jev to select words or verify locally drafted continuations.
Real data. Real impact.
Emerging
Developers
git clone https://github.com/y0usaf/jev-lm.gitPer week
Excellent
One command unlocks capabilities that would take hours to build.
A word-level generation experiment that asks Jev to select words or verify locally drafted continuations.
Choice selects the next word; Noul evaluates candidate chunks and stopping conditions.
Exposes the limitations of using a decision model as a text generator.
A research prototype whose reported small-vocabulary evaluation did not beat a unigram baseline; not a general language-model replacement or independently retested here.
Jev is TypeSafe AI's System One decision model: it returns typed scores and choices (not free text), so agents can route, filter and gate at low latency and cost.
Source: y0usaf/jev-lm — 6 stars, MIT. Listed by Awesome Jev.
MIT
4,600+ AI skills, agents & workflows. Install in 60 seconds. Part of the Torly.ai family.
© 2026 Torly.ai. All rights reserved.