User repository on GitHub
Is this headline satire or real news? Ask Jev and AnyJev side by side.
- Forks
- 1
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Demos & Experiments
Onion or Not
A small Python app that asks whether a headline is Onion-style satire or real news. Jev runs through TypeSafe's API with your own key and is compared side by side with AnyJev L0 and raw Qwen3-8B running locally, with an option to swap the answer order.
- Stars
- 9
- Language
- Python
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- andrisgauracs
- Last verified
- Oct 2026
- Languages
- Python
- Capabilities
- Open sourceRuns locallyRequires API key
Highlights
- Two open models run locally through llama.cpp while Jev answers from TypeSafe's API with your own key.
- The option-swap toggle asks both satire-or-real-news and real-news-or-satire to test answer flipping.
- AnyJev L0 reads both orders, so its answer cannot flip the way a plain LLM's sometimes can, per the README.
- Quick start is uv sync, copy .env.example to .env with a TypeSafe key, then uv run onion-or-not download for the model.
- MIT licensed, written in Python for Python 3.10+ and uv, created and pushed 2026-10-16.
Quickstart
bash
git clone https://github.com/andrisgauracs/onion-or-not.git
cd onion-or-not
uv sync
cp .env.example .env
uv run onion-or-not downloadWatch out
README reviewed 2026-10-17; the three-way setup needs both a TypeSafe key and about 5 GB of local model weights, and the README excerpt shows no accuracy results, only the comparison method.
Questions about this entry
Short answers drawn from the linked project or page.
A small demo app that asks whether any headline is Onion-style satire or real news. Type a headline and three models answer side by side, each with how sure it is: Jev, AnyJev L0 and raw Qwen3-8B.
Jev is TypeSafe's hosted decision model, called through TypeSafe's API with your own API key. The other two run on your computer through llama.cpp: AnyJev L0 (Qwen3-8B read out by AnyJev at level L0) and the raw Qwen3-8B model.
It asks both satire-or-real-news and real-news-or-satire. A plain LLM sometimes changes its answer just because the options came in a different order, while AnyJev L0 reads both orders so its answer cannot flip that way, per the README.
Python 3.10+ and uv, a TypeSafe API key, and the ability to download about 5 GB of open-model weights for the local runs. The quick start is uv sync, copy .env.example to .env with your key, then uv run onion-or-not download.
The digest shows the setup sequence up to uv run onion-or-not download for the model; how the app itself is launched is not shown in the excerpt, and the only visual is the README's docs/screenshot.png reference.
No accuracy or calibration results appear in the digest excerpt; the README explains the three-way setup and the order-swap control rather than publishing scores, so its comparison is a demo, not a measured evaluation.
The GitHub user andrisgauracs publishes it under the MIT license. The repository was created and last pushed 2026-10-16, with 4 stars and 1 fork when the sweep ran.
More Jev resources like this
An open-source browser 3D character split between two models: GPT-Live-1 talks with you over WebRTC while Jev reads the live transcript and picks Annie's face, gestures and body language in about 120 ms, with code gates on every answer.
Jev Demos & Experiments#community#github#python
Community
rehan-remade/open-anniePython
rehan-remade/open-anniePythonEleven small runnable projects for learning Jev and other System One decision models, each lesson pairing a one-command web demo with a whiteboard and README, from the three question types through jobs sorting and contract review to Jev inside a SQL query. Every demo makes live API calls.
Jev Demos & Experiments#community#github#python
Community
aianytime/jev-crash-coursePython
aianytime/jev-crash-coursePythonFive self-contained Jev demos, one folder each: Jev fetching for Claude, Jev and Laya grading 500 AI answers, a prompt-injection review gate, and two fine-tuned decider comparisons against Jev.
Jev Demos & Experiments#community#python#demos
Community
mani-aiml/jev-demosPython
mani-aiml/jev-demosPythonA small proof-of-concept app that runs local Laya and external System One-family APIs in one judgment UI: it executes typed questions against a state and compares choice probabilities, yes/no probabilities and ordinal scores across providers.
Jev Demos & Experiments#community#github#python
Community
keispace/systemonePython
keispace/systemonePythonA Python dispatcher skeleton for TypeSafe System One (Jev) that turns typed signatures into batched questions, plus a Playwright browser harness that measures the dispatcher against ground truth.
Jev Demos & Experiments#community#github#python
Community
monkey1sai/jev-systemone-labPython
monkey1sai/jev-systemone-labPythonA Streamlit lab that places a bounded System One decision inside a coding-agent workflow: Laya classifies CI failures, LangGraph applies a fallback policy, and an OpenAI agent inspects a fixture repo and proposes a fix.
Jev Demos & Experiments#community#github#python
Community
natenberenstein/systemonePython
natenberenstein/systemonePythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.