Heman10x-NGU/openJev-verdict-2.0

Calibrated 151M Non-Autoregressive Decision Engine beating TypeSafe Jev & Laya on LocalLLaMA/typed-decisions (77.10% acc, 0.0636 Brier, 0.0144 ECE)

Language modelsPython#agentic-ai#brier-score#calibration#decision-engine#deep-learning

5.0 Early signal Magnitude out of 10. Star growth looks organic. Data as of September 20, 2026.

Open on Seismograph Open on GitHub

About the project

An open non-autoregressive decision model built on ModernBERT (149.6M) that emits typed decisions (Choice, Score) with calibrated probabilities in ~20-25 ms. On the typed-decisions benchmark it reaches 77.10% accuracy at 1.44% ECE.

Useful for

Embed as a gating layer that auto-accepts confident decisions and escalates uncertain ones to humans
Replace parsing of LLM text outputs with a direct model call in an automation pipeline
Run the verdict2.evaluate benchmark on your own data to check calibration

README summarized by DeepSeek V4.1 Flash. Details may be inaccurate.

Why it’s trending

12 stars so far today, about 59 expected by the end of the day.
The spike has held for 2 days in a row — not a one-off blip.
The repository is 1 day old and already has 98 stars.
Top new repositories this week: #148.
About 17 forks a day — people are taking the code.

Stars per day

Bars are daily stars, the line is the usual pace. Red marks spike days.

Numbers

Total stars: 98
Stars in a day: 59
Forks: 12
Issues and pull requests: 1
Watchers: 0
Language: Python
Created: September 19, 2026
Last push: September 19, 2026

Star trust

Growth looks organic: forks and discussion are in line with active projects, and stars arrive unevenly, the way people give them.

These are heuristics, not a verdict: we judge by the repository’s behavior, not by a list of stargazers.

Spotted in

September 20, 2026Top new repositories this week: #148

README badge

Paste it into your README — the badge shows the current magnitude and links to this page.

[![Seismograph](https://gitnova.dev/badge/Heman10x-NGU/openJev-verdict-2.0.svg?lang=en)](https://gitnova.dev/en/r/Heman10x-NGU/openJev-verdict-2.0)

Similar projects

7.2
TianyuCodings/NanoJev
A nano replica of Jev built on Qwen3-0.6B that outputs probability distributions over candidates in a single forward pass with no token decoding, plus a training pipeline and game demos (maze, Snake).

Early signalLanguage modelsPython+316 stars in a day
7.0
bespokelabsai/nimble
Nimble is a model and training recipe for fast typed decisions over text: given a flat schema of enum and boolean fields, it picks an answer and returns probabilities for each option. It targets routing, condition…

Early signalLanguage modelsPython+318 stars in a day
7.0
higgsfield-ai/higgsfield
Fault-tolerant GPU orchestration and ML framework for distributed training of models with billions to trillions of parameters, including LLMs. Manages node access, experiment queueing, and GitHub Actions integration.

BreakoutLanguage modelsJupyter Notebook+220 stars in a day
6.9
githubnext/localjev
A local TypeScript/Bun HTTP service implementing a Jev-compatible POST /v1/systemone API on top of an OpenAI-compatible Chat Completions endpoint (DiffusionGemma via oMLX). It acts as a bridge for typed decisions…

Early signalLanguage modelsTypeScript+239 stars in a day
6.8
docling-project/docling
Docling is a document parsing library for many formats (PDF, DOCX, PPTX, XLSX, HTML, EPUB, audio, video) with advanced PDF understanding: page layout, tables, formulas, OCR. It prepares structured data for generative…

BreakoutLanguage modelsPython+301 stars in a day
6.7
jaredpalmer/kev
kev is a LoRA adapter with a small readout head on top of Qwen2.5-0.5B that answers many typed questions about a document in a single forward pass, returning calibrated probabilities instead of text.

Early signalLanguage modelsPython+223 stars in a day