User repository on GitHub
Typed, calibrated decisions from a local model. No text generated.
- Forks
- 2
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Models & Reimplementations
Ruling
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.
- Stars
- 10
- Language
- Python
- License
- MIT
- Published by
- Community
- Last verified
- Oct 2026
About this project
Typed decisions from a local model, no text generated. Not affiliated with TypeSafe AI. ruling turns any open chat model into a decision engine like TypeSafe's Jev. You send a state and typed questions; you get back a probability for every option you declared. Nothing is generated, so there is nothing to parse and the model cannot invent an answer you did not list.
Quoted from the project's README. Read the full README.
- Languages
- PythonTypeScript
- Platforms
- CLIAPImacOSOpenRouter
- Capabilities
- Open sourceSelf-hostedRuns locally
Highlights
- Default model is mlx-community/Qwen3.5-4B-4bit at 2.3 GB; three questions take about 136 ms on a Mac.
- On 256 public judgments carrying Jev's own answers, a 19 GB Qwen3.6-35B-A3B scored 231 to Jev's 238 (McNemar p = 0.21).
- Includes optional LoRA fine-tuning, holdout-based eval, temperature calibration and cascade to a secondary model.
- Ships Python and zero-dependency TypeScript clients; TypeSafe's official SDK works via TYPESAFE_BASE_URL.
Watch out
The fast MLX path is Apple Silicon only, one request runs at a time, and it handles text only with no images or audio. MIT licensed and independent from TypeSafe AI.
Questions about this entry
Short answers drawn from the linked project or page.
No. Serving works with stock MLX checkpoints; an optional LoRA fine-tune on a few hundred of your own labeled decisions improves calibration when you want thresholds to be safe to act on.
No. It is an independent MIT-licensed implementation that reuses the /v1/systemone request shape so the official SDK and other clients can point at it unchanged.
The fast MLX path needs an Apple Silicon Mac, and the quickstart uses Python 3.12 or newer with uv; its first serve downloads a 2.3 GB model, and one request runs at a time.
Set a secondary model with RULING_CASCADE_TO and a RULING_CASCADE_THRESHOLD; the secondary answers only questions whose top probability falls under the threshold, and the pair can match the secondary alone.
Both directions survive an outage: if the secondary returns 503 or times out, the primary's answers stand for the questions it would have escalated, and RULING_FAILOVER_TO handles the reverse case.
Yes. The trainer refuses to start if any training state also appears in a holdout file, so leakage cannot happen by accident.
No, it is text only with no image or audio input; separately, a calibration file it produces is bound to the exact model and adapter it was fit on and will not load against anything else.
Related terms
Glossary definitions related to this entry.
More Jev resources like this
130GitHub 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/autojevPython3GitHub stars
A training-free runtime that turns dense Qwen3 or Qwen3.5 checkpoints into a Jev-style Noul, Choice and Score service, sharing one state prefill across isolated questions and reading each answer from candidate-token logits.
Jev Models & Reimplementations#community#qwen#local-inference
Community
HopLee6/QwevPython
HopLee6/QwevPython77GitHub stars
An Apache-2.0 0.8B decision model that returns calibrated probabilities for yes/no, choice and score questions on CPU. Distributed as GGUF for llama.cpp and served behind a Jev-style /v1/systemone API, so existing Jev clients mostly need only a new base URL.
Jev Models & Reimplementations#community#gguf#local-inference
Community
kouhxp/gutsyHTML
kouhxp/gutsyHTML17GitHub 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/vevPython877GitHub 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-flowPython109GitHub 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-JevHTMLFrom 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.