GitHub · Benchmarks and evaluation · Runtime ports

Heman10x-NGU/openJev-verdict-2.0

A second opinion on small decision models, with the caveats attached.

The repository's GitHub card, read on 2026-09-24

What it does

Trains a general-purpose decision model on a small encoder with a classification head, publishes the checkpoint on Hugging Face, and ships an in-browser engine so it can answer without a server.

The repository's own description: “Calibrated 151M Non-Autoregressive Decision Engine beating TypeSafe Jev & Laya on LocalLLaMA/typed-decisions (77.10% acc, 0.0636 Brier, 0.0144 ECE)”

How it works

The public checkpoint is a trained head over ModernBERT-base, and its comparison table is measured on a benchmark the repository names — JevBench — which is not the benchmark any other row in this column reports.

The repository, by the numbers

Stars

283

Licence

Apache-2.0

Stack

Python · ModernBERT-base · GLiClass head

Last push

2026-09-20

Read from the GitHub API on 2026-09-24. Stars and the last push move daily — quote them with the date, the way we do.

What we checked

the README, read 2026-09-24
the accuracy and calibration badges on the repository page
the note that two different models live in the repository
the licence file

What we did not check. Everything in its comparison table. We have not run the model, we had not seen JevBench before this repository, and a number on an unfamiliar benchmark is a claim rather than a fact.

Open the repository ↗All builds

The rest of the topic

Jev, the reference →

Start here

Laya, the reference What it is, what we measured running it, and what it is not.
Laya and Jev, compared The two models on the same questions, including the ones Laya loses.

Columns

Built with Laya 18 you are here Ports, servers, agents and games, with what we checked.
What X is saying 22
What people filmed 9
What Hacker News made of it 6
Docs, weights and reading 15

The topic

Every record, by kind of work The same rows cut a second way, across every column.
What self-hosting costs Hardware, memory, and the failure modes of running it yourself.
What the licence allows Apache-2.0 weights, three checkpoints, and the limits that come with them.

By kind of work

Runtime ports 23 Getting the weights onto someone else's hardware: MLX on a Mac, Core ML on the Neural Engine, ggml in C++, candle in Rust, ONNX in Node.
Serving and API 8 Standing the model up as something else can call, usually behind Jev's own wire format.
Agents and tools 6 Typed decisions inside an agent loop: routing, evals, guardrails, memory, and the plumbing around them.
Browser and computer use 2 Deciding the next click, key or field on a real screen.
Games and demos 9 The model playing something you can watch. Most of this model's traffic arrived through one of these.
Benchmarks and evaluation 19 Numbers somebody actually ran — including the ones that say Laya lost.

What it does

How it works

The repository, by the numbers

What we checked

More like this

mizorewww/laya-mlx

mizorewww/laya-coreml

receptron/laya