User repository on GitHub
A 25-second IT buzzword game to feel how fast and cheap Jev (TypeSafe System One) is — code + Jev only, no LLM.
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Demos & Experiments
Jev Buzzword Rush
A 25-second IT-buzzword typing game built to feel how fast and cheap Jev is: each word gets five typed judgments in one request, and the screen shows the running cost next to the real HTTP request and response.
- Stars
- 1
- Language
- JavaScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- pochang6
- Last verified
- Oct 2026
- Languages
- PythonJavaScript
- Platforms
- Browser
- Capabilities
- Open sourceRuns locallyRequires API key
Highlights
- Five typed questions per word in one request: IT-field Choice plus activity, learning cost, connoisseur and trivia scores.
- Measured demo run: 14 judgments at about 400 ms average, clearing the game for roughly ¥0.14 at $1 = ¥160.
- Python 3.10+ with the standard library only — no pip install — plus a modern browser with Intl.Segmenter.
- Retorts are chosen in code from prepared lines based on Jev's judgments; no model generates any text in the game.
- The on-screen API log shows live requests with the key masked as Authorization: Bearer ********.
Quickstart
bash
git clone https://github.com/pochang6/jev-buzzword-rush.git
cd jev-buzzword-rush
TYPESAFE_AI_KEY=xxxxxxxx python3 server.pyWatch out
Copy reviewed against the truncated repo README on 2026-10-15, including its English section; the excerpt stops inside the example question JSON. The ~400 ms and ¥0.14 figures come from one demo run per the README, which also notes that issuing a TypeSafe key requires a credit purchase of at least $5.
Questions about this entry
Short answers drawn from the linked project or page.
A small game built to feel how fast and cheap Jev is in about 25 seconds: type IT terms and Jev judges each one as you go. There is no submit button — each word is judged as it is typed, and the meters fill per field until all five are full.
Because feeling Jev's speed is the point: if an LLM generated the retorts, perceived speed would be gated by the LLM's response time. The game generates no text at all — Jev judges, and code picks prepared lines based on those judgments.
One request carries five questions per word: a Choice for the IT field (cloud, network, database, programming, security, or not IT) plus scores such as how current the term is, learning cost, connoisseur-level and which trivia line fits.
Clone the repo and run TYPESAFE_AI_KEY=xxxxxxxx python3 server.py, then open http://localhost:8787. /?demo auto-types the script one character at a time, and /?record shows a button that records the tab and downloads a webm.
The README's measured demo cleared the game with 14 judgments at about 400 ms average for roughly ¥0.14 at $1 = ¥160. The key comes from console.typesafe.ai, and the README notes issuing a key requires buying credit, minimum $5.
Python 3.10+ using only the standard library — no pip install — and a recent browser with Intl.Segmenter, such as Chrome. The TypeSafe key is passed via the TYPESAFE_AI_KEY environment variable and never enters the repository.
web/config.js holds the game knobs: perCategory 2, threshold 0.8 for what counts as an IT term, maxCandidates 8 (five questions per word), askRate 0.6, idle and chunk limits, the demo script and 15 trivia lines. Fields, retort wording and the bear's expressions live there too.
More Jev resources like this
A 12,000-neuron fly-brain subgraph of the maleCNS connectome fights Jev in mk.js: the fly simulates spikes in the browser at 10 decisions per second while Jev reads the fight as JSON and answers one typed choice among seven moves.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
lavallee/mk-jev-fly-brainJavaScript
lavallee/mk-jev-fly-brainJavaScriptA minimal real-time Clash Royale-style game engine in Node.js that pits a Jev System One bot against a traditional Ollama LLM bot, with an MCP server and a canvas web client. The Jev bot answers four tactical questions in one parallel call.
Jev Demos & Experiments#community#catalog-import#logicrw
A workshop of small, inspectable Python experiments for TypeSafe System One models: ten critical bays, from fraud screening to emergency dispatch, each giving Jev a realistic state while deterministic policy keeps authority over the next action.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
jgalego/jevs-garagePython
jgalego/jevs-garagePythonA table-tennis game where every paddle movement on the right is a live Jev decision: the relay renders the state as prose, asks one multiple-choice question, and the option label IS the pixel value. No local prediction, no fallback.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
liuhao-1443/jev-table-tennisPython
liuhao-1443/jev-table-tennisPythonPlay Chinese Chess against Jev: every AI move fans out a Score 0-4 question per legal move in a single system_one() call, then plays the highest-scoring move. No minimax and no eval tables, just Jev's read of the position, with the top 10 shown as score bars.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
zafer-liu/jev-xiangqiHTML
zafer-liu/jev-xiangqiHTMLA small local app that turns Jev's per-pixel probability distributions into paintings: stroke pigments are sampled from the probabilities, mean color supplies the underpainting, and entropy controls relief, rendered in a browser worker.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
achimala/jevinciJavaScript
achimala/jevinciJavaScriptFrom 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.