Docs · Guides & docs
Score
Rate content against 2 to 10 ordered levels.
Open docs.typesafe.ai ↗How builders describe it
Rate content against 2 to 10 ordered levels. Returns a probability-weighted score that can land between levels, per-level probabilities, and confidence, with the arithmetic left to code.
The decision Jev makes
Where to start: the quickstart, the decision primitives and the patterns.
Where it fits
The official starting points plus the best explainers, videos and articles about Jev and RLCD. All 231 guides & docs projects →
Related projects
Guides & docsJEV explained in 17 minutes with codecodebasics explains Jev with runnable code.YouTube · youtube.com/@codebasics
Guides & docsJev explained with Python examplesZazenCodes works through Jev hands on in Python, from a first call to typed decisions.YouTube · youtube.com/@ZazenCodes
Guides & docsJev is the first of a whole new class of AI modelsCole Medin on what the class is, and how to actually use it.YouTube · youtube.com/@ColeMedin
Guides & docsJev vs. Laya: what they are and the controversyAdam Gardner on both decision models and the argument around them.YouTube · youtube.com/@AdamGardner
Guides & docsJev’s Architecture UnmaskedArcher Hume probed Jev with 10,000 API calls to work out roughly how it is built.Guide · archerhume
Guides & docsRun a Jev-style model on your own GPUCloud Codes sets up a Jev-style decision model on local hardware.YouTube · youtube.com/@CloudCodes