Organization repository on GitHub
Jebadiah, an open System One decision model: trainer, data builders, evals and every run record
- Forks
- 1
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
CommunityJev Models & Reimplementations
Jebadiah
Open System One decision models (27B, 9B and 4B on Qwen chat bases, in bf16, GGUF and MLX) that answer Choice, Noul and Score questions with a probability for every option from one forward pass. The repo carries the trainer, data builders, evaluation harnesses and a record of every run.
- Stars
- 5
- Language
- Python
- License
- APACHE-2.0
- Published by
- Community
- Last verified
- Oct 2026
About this project
Jebadiah (Jeb for short) is an open decision model. It answers small typed questions about a piece of structured state, and instead of writing text it returns a calibrated probability over the allowed answers. This repository holds everything needed to reproduce it: the trainer and evaluator, the data builders, the evaluation harnesses, the configuration of every run, and the result record of every run we report.
Quoted from the project's README. Read the full README.
Highlights
- Published jebadiah-27b scores a 78.95 headline; 9B v2 reaches 73.93 and 4B v2 72.49.
- LoRA rank 16, alpha 32, dropout 0.05; 32.5M to 116.7M trainable parameters; one epoch, seed 17.
- The 27B run trained on one H200 in 167 minutes; weights live on Hugging Face under frontier-infra.
- Serves through AINode's /v1/decide and /v1/systemone routes; every run record is committed.
Watch out
Apache-2.0. Results are self-reported single-seed runs without significance tests, and serving currently relies on AINode.
Questions about this entry
Short answers drawn from the linked project or page.
It reads its answer from candidate label-token logits at one position in a single forward pass, returning calibrated probabilities over allowed options instead of generating text, so there is no generation, parsing or refusal path.
Published on Hugging Face under the frontier-infra organization as one collection covering the 4B, 9B and 27B versions; the repository holds the trainer, configurations and run records needed to reproduce them.
With AINode's trainer on public data: LoRA rank 16, alpha 32, dropout 0.05, giving 32.5M to 116.7M trainable parameters, one epoch at seed 17; the 27B run took 167 minutes on one H200.
jebadiah-27b scores a 78.95 headline, 9B v2 reaches 73.93, and 4B v2 72.49, all measured with the repository's own evaluator on public sets.
Yes. The repository ships the trainer, data builders, evaluation harnesses, verbatim prompt copies, and the configuration and result record of every run, including the ones not published.
The repository and weights are Apache-2.0, but the published results are self-reported single-seed runs without significance tests.
Public, license-checked sources only: LocalLLaMA/typed-decisions, the Kev v7 training sources, HelpSteer2 train and SummEval, with every evaluation item excluded and every revision and license recorded.
More Jev resources like this
109GitHub stars
Open decision models from a self-improvement loop that publishes every experiment, failures included. They answer Noul, Choice and Score questions on the same POST /v1/systemone wire format as Jev, returning a calibrated probability per option in one forward pass; the 27B scores 65.64 on the Decision Index 0.3 set.
Jev Models & Reimplementations#community#decision-models#system-one
Community
Shanghua-Gao/RSI-JevHTML
Shanghua-Gao/RSI-JevHTML130GitHub stars
An open recipe for a 27B multimodal decision model trained with full-weight SFT on 73,000 examples over one H200, serving choice, noul and score on /v1/systemone with per-checkpoint provenance and released calibration plots.
Jev Models & Reimplementations#community#calibration#qwen
Community
denis-pplx/autojevPython
denis-pplx/autojevPython17GitHub stars
Jev-compatible decision models with vision input: LoRA fine-tunes of Qwen3.5 at 4B and 9B serving Choice, Score and Noul on the /v1/systemone wire format, with screenshots and photos placed beside text in the state. Runs on your own NVIDIA GPU; weights are non-commercial.
Jev Models & Reimplementations#community#vision#open-weights
Community
Xiaooolong/vevPython
Xiaooolong/vevPython10GitHub stars
A local Jev-compatible server that reads Choice, Score and Noul answers from the logits of any MLX checkpoint or OpenAI-compatible endpoint with no training, serving POST /v1/systemone as a drop-in for the official TypeSafe SDK.
Jev Models & Reimplementations#community#mlx#calibration
Community
bradAGI/rulingPython
bradAGI/rulingPython877GitHub stars
A local server that reads next-token probabilities from a fine-tuned Spark model and returns Choice, Score, and Noul answers with zero generated tokens.
Jev Models & Reimplementations#community#python#open-models
Community
Rizzo-AI-Academy/rizzo-flowPython
Rizzo-AI-Academy/rizzo-flowPython58GitHub stars
A local /v1/systemone server that reads label probabilities from a GGUF chat model, with prompt repetition and a confidence shrink toward uniform.
Jev Models & Reimplementations#community#python#open-models
Community
Mushroom-Systems/lichenPython
Mushroom-Systems/lichenPythonFrom the Jev guides
Original write-ups on what this entry is for.
Jev takes text only; Clef-omni adds image, audio, and video input and Liquid d1 adds images, each keeping the same typed-answer shape. Where that matters, where it does not, and the costs the vendors underplay.
Jev Models & Reimplementations#multimodal#open-models#decision-model
Read article
One model in September, a dozen by October. Clef, Liquid d1, Perplexity, and OpenAI all ship the same typed-answer shape now. How to read the rival benchmarks and pick on your own task.
Jev Models & Reimplementations#decision-model#benchmarks#pricing
Read article
Hundreds of repositories carry the system-one tag, but the ecosystem is a dozen ideas in different sizes. A tour of the families, the licenses, and what a download does not include.
Jev Models & Reimplementations#open-models#hugging-face#local
Read article
From the Jev community
Posts from builders shipping with Jev right now.