User repository on GitHub
Write a fighter in plain English. Jev pilots it in real time. PR-driven ladder, swappable brains.
- Forks
- 0
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Demos & Experiments
Jev Arena (plain-English fighters)
Fighters in this arena are paragraphs of plain English, and Jev pilots them in real time — typed move, action and reflex decisions with live probabilities. Join the ladder with a pull request and swap the brain: Jev, Laya, kev or any LLM.
- Stars
- 1
- Language
- JavaScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- eliot5566
- Last verified
- Oct 2026
Used inJev for Gaming
- Languages
- JavaScript
- Platforms
- Browser
- Capabilities
- Open sourceRequires API key
Highlights
- A fighter is a YAML block: name, author, colour, a strategy paragraph, optional reflexes with thresholds and a fallback — no code or coordinates.
- One request asks how to move and what to do as Choices plus each reflex as a Noul; a fired reflex wins its slot, else the top choice, else the fallback.
- Zen, a one-sentence fighter with zero reflexes, finished 2nd of 6 at 1062 Elo, one point behind Trickster; the carefully tuned Sniper went 0-10.
- The six-fighter ladder ran on real jev-1.13.0 over 30 fights that all ended in KO; every replay and the full table are in ladder/LEADERBOARD.md.
- TypeSafe's numbers put Jev's typed decisions at roughly 70 to 500 ms — fast enough, the README argues, to be a fighter's reflexes while a chat model writes.
Watch out
Grounded on the truncated README and GitHub metadata read on 2026-10-15; the capture ends right after the How one decision works section, so the quick-start commands and the pull-request ladder mechanics are not covered here. The Elo table comes from the project's own 30-fight run, and the 70–500 ms timing is attributed to TypeSafe rather than measured in the repo.
Questions about this entry
Short answers drawn from the linked project or page.
A real-time bot battler where you write the strategy in English and a System One model supplies the reflexes live, several times a second. The default brain is Jev, and it can be swapped for Laya, kev or any LLM.
A YAML block: name, author, colour, a strategy paragraph, a list of reflexes each with a when, a do and a threshold, and a fallback. The README's Tortoise stays near a pillar, shields when a heavy shot is on its way and answers with bolts; there is no code.
The engine computes the geometry, then one parallel request asks how to move and what to do as Choice questions plus each reflex as a Noul. A reflex fired above its threshold wins its slot; otherwise the top choice is used if confident; otherwise your fallback move or wait.
On the current Jev ladder Trickster leads at 1063 Elo over 8-0-2, one point ahead of Zen, a one-sentence fighter with no reflexes. Tortoise is third in the six-fighter table; the tuned Sniper is last at 0-10. The full table and replays are in ladder/LEADERBOARD.md.
Zen's whole strategy is Win the fight. Do whatever a wise, calm fighter would do in this exact moment. The README presents its 2nd-of-6 finish as the open question of how far one sentence goes, and invites you to write a paragraph that beats it.
The ladder is PR-driven — you join with a pull request — and the README links the live ladder on GitHub Pages where you can watch it; the repo is also published on npm as jev-arena for playing locally. The captured README ends before detailing the submission steps.
The README's ladder ran on real jev-1.13.0, in real time, over 30 fights; the captured section does not document key setup or hardware requirements, and the 70–500 ms figure is TypeSafe's own, not a measurement in the repo.
More Jev resources like this
Jev and GPT-6 Luna race through the same Flappy Bird course, with every flap a live model decision. Both birds request flap or wait every 50 ms, and the browser shows each answer applied, superseded or skipped.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
jaibhasin/jev-flappy-birdJavaScript
jaibhasin/jev-flappy-birdJavaScriptA 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 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-tennisPythonA playable Three.js driving simulator where Jev-powered autopilot picks steering and speed up to four times per second: compact tables of eligible paths, traffic and road boundaries go to Jev while geometry and control math stay local.
Jev Demos & Experiments#community#catalog-import#cobanov
Community
standardagents/jevpilotJavaScript
standardagents/jevpilotJavaScriptA Three.js first-person garden game where a glowing fly hunts goblins and TypeSafe's Jev steers: about five times per second it answers four typed questions on turning, sidestepping trunks, climbing and diving, and every flight is logged.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
webdevcody/jevs-flyJavaScript
webdevcody/jevs-flyJavaScriptA public operator lab for TypeSafe Jev: a chess match as the flagship page, a row of smaller judgment booths, and one shared client that keeps the API key out of the browser via a local Python server or a Cloudflare Worker.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
4anti/jev-broadcast-labJavaScript
4anti/jev-broadcast-labJavaScriptFrom 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.