Skip to content

Decis

A self-hosted System One server: one /v1/systemone contract, a Docker image each for Laya and kev, and a playground where snake, dino, and tetris call that endpoint.

Category
Models & Reimplementations
Format
—
Published by
Community
Author
chaitin
Added
2026-09-28
Last verified
2026-09-28

Highlights

  • The official typesafe-sdk talks to it by changing base_url. The wire contract is /v1/systemone.
  • Laya and kev each ship as a multi-arch image with weights included. The default engine is laya-multilingual.
  • Snake, dino, and tetris each decide through a real /v1/systemone call. The engine listens on port 8000 and the games on 8080.
  • Status is v0.3.2, pre-1.0. /healthz answers immediately. /readyz stays on loading until the model can answer.

Quickstart

bash
docker compose -f docker-compose.yml up -d --wait

Watch out

Apache-2.0. Pre-1.0: the v1 wire contract only adds fields. Needs Docker and a DECIS_API_KEY in .env. Distinct from the hosted Jev games and from the Laya and kev model repos themselves.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

Hacker News: Show HN: a mouse colony where each next move is Jev or a fixed rule.

1 pointShow HN: Jev-mice – mouse colony simulation using Jev and deterministic engineRead the thread on Hacker News

X: Jev plus Astra beats the Ender Dragon for under a dollar

X: Jev plays Pokémon Showdown

Related terms

Glossary definitions related to this entry.

More like this

1.1kGitHub stars
A local yes/no decision model that speaks Jev's wire format on a laptop CPU, refusing Choice and Score until those question types ship.
Models & Reimplementations#community#python#open-models
689GitHub 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.
Models & Reimplementations#community#python#open-models
57GitHub stars
An Apache-2.0 decision engine with a TypeSafe-compatible endpoint, document answers with source passages, and a ViZDoom demo that picks actions from structured game state.
Models & Reimplementations#community#python#local
Back to all resources

From the community

Posts from builders shipping with Jev right now.

Follow @typesafeai

A local Telegram analyzer with Jev as the classifier

I vibe-coded this simple, open-source and local Telegram content analyzer using Jev as a classifier. Jev Classifier • jevclassifier.vercel.app Export your channel data as JSON and it will analyze each post's intent, quality, sentiment, and reaction tone and if it's a DM/Group Show more

Diogo Almeida
Diogo Almeida
TypeSafe AI
@CompleteSkeptic

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x

Reply

A Krunker-inspired FPS, decided by Jev

Maybe I’m just washed, but Jev is beating me at my own game. Built a Krunker-inspired FPS where you 1v1 Jev. Try it out yourself: jev-arcade.vercel.app/duel

Diogo Almeida
Diogo Almeida
TypeSafe AI
@CompleteSkeptic

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x

Reply

Pixel art out of 256 multiple-choice questions

I turned @typesafeai Jev model into a pixel art model. 256 multiple choice questions, one per pixel, plus refinement passes. These were the best results out of the bunch I tested. Code: github.com/Wizhill05/type…

Image
Image
Image
Image
Diogo Almeida
Diogo Almeida
TypeSafe AI
@CompleteSkeptic

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x

Reply

An Elixir port of the official SDK

jod: Zod validates data, Jev judges it

Loki: a self-improving harness with a Jev companion