Ra mắt ngày 25 tháng 9, 2026

AI-assisted Windows and Chrome GUI delegation for Codex: task contracts, local execution, Jev semantic decisions, recovery and outcome verification.

Claude Code skill: hand bulk judgment calls to Jev. 86% fewer Claude tokens on a 12-task benchmark, up to 20x faster. One-line npx install.

High-performance, high-precision, locally-fast-deployed Jev-class medical decision service on national medical exams — four readings, one routed decision, zero training.

Give your agent eyes: any video link in, one typed JSON out. A Claude skill + API.

Fast decision routing for agent harnesses — native MCP with Jev for tool selection and MCTS for multi-step decisions.

An open System One decision model

Small, calibrated decision models on your own machine: systemone-compatible local server, 6 agent skills, Claude Code guard, MCP tools. Weights on Hugging Face.

Jev-LCT: Open System-One Decision Engine with Free Calibrated Confidence from Recurrent Trajectories

Emoção da fala em tempo real: transcrição ao vivo com Gemini 3.5 Transcribe Live e seis emoções avaliadas pelo Jev (TypeSafe) a cada 0,5 s. Projeto aberto da illumi.

Browser driving game that captions dropped road images locally, then uses Jev Choice answers to drive the car while code owns physics and timing.

System 1 for your coding agent: cut token use by deciding, not generating.

Shared memory Jev-based plugin for multi-agent coding: subagents and Codex/Claude Code/Cursor hand-offs share one project memory. 一个多agent协同高效记忆统筹插件

Ultra-fast test impact analysis powered by Jev. Run only the tests that matter.

JeVerifier: cheap Jev (TypeSafe) checks that keep code maintainable and docs consistent, plus context retrieval — modest token savings

Pi extension: auto-approve routine Bash commands with a System One (Jev) decision model, escalate the rest to a human

A library to chain Jev calls, for the deranged and enlightened

Cross-agent memory explanation skill powered by sivtr

Hermes Agent plugin: TypeSafe Jev model routing + trim-then-compress

Probabilistic control flow for TypeScript - powered by TypeSafe's Jev

Write a fighter in plain English. Jev pilots it in real time. PR-driven ladder, swappable brains.

Open-source AI agent decision layer for Codex, Claude Code, Hermes, Antigravity and VS Code. TypeSafe Jev MCP routing, 38 recipes, local gates and receipts; optional Laya-MLX on Apple Silicon.

Async, typed intent & tool routing for Python agents: Jev decides, a confidence gate stops weak decisions before they reach a privileged tool.

Grounded information extraction that cannot hallucinate: code proposes spans, Jev decides. An open-source alternative to LangExtract, with a bilingual benchmark and paper.

Auditable Jev decisions through OpenRouter. A macOS-first TypeScript CLI with Keychain credentials, recorded costs, and conservative retries.

LISA: Leak, Injection & Simplicity Auditor. GitHub Action that uses TypeSafe Jev (a system one model) to flag secrets, security vulnerabilities, and unneeded complexity in pull requests

Fast, calibrated, staged code & security review (momus)

One Go binary that runs low-cost AI coding agents with a judge on every action. Known mistakes are caught in the act.

Experimental plugin to convert protobuf into Jev code

Seems: Python, plus judgment, answered locally by Laya. A variant of kavehmz/seems-lang with the open-weight decision model swapped in for TypeSafe Jev.

Daily AI blog digest: Feedly OPML -> Jev -> LLM summaries -> Telegram

For people with aphasia: describe the word you can't find, tap the right guess, hear it said. It only picks from a fixed word list plus your own names, using Jev by TypeSafe.

Autonomous anti-spam and moderation engine for WhatsApp groups powered by TypeSafe Jev System-1 judgment primitives and zero-latency two-tier defense.

Projet expérimental pour un agent de bureau de recherche, classement et résumé de documents locaux; Jev à l’étude pour les décisions structurées.

Bobcat: a typed-decision model. State in, Choice / Noul / Score out, with a probability for every answer you name; no generated text. Compiler, server, training and evaluation code for Bobcat 1.1 and Bobcat Flash 1.1.

Claude Code: Jev tool gate + Anthropic-compatible proxy (route + trim→compress)

Use and fine-tune System One models (Jev, Laya) on your data, with an LLM as the teacher.

DSH 插件 · 用 Jev（TypeSafe System One）判定推理强度：默认只切思考强度（零缓存代价），自动模型路由出厂关闭 + 硬门禁 + 成本闸。作者 chenshi.ai

个人用jev

Demo of how to use JEV — includes an example iPhone 16 review dataset for testing/practice.

voice assistant that drives your desktop

별빛 검문소: Korean mobile document inspection game with real precomputed Jev decisions

JEV Studio is a small Next.js evaluation lab for turning natural-language input into structured signals with the Typesafe SystemOne API.

Per-turn reasoning effort for coding agents, classified by TypeSafe's Jev via AI SDK (Vercel AI Gateway / OpenRouter / TypeSafe). Claude Code plugin; Codex, Cursor, Amp next.

Sort your Gmail inbox into categories you write in plain English. TypeSafe's Jev makes each call with calibrated probabilities; uncertain mail gets a Review label. Built on One, runs with zero keys.

Native process orchestration with typed, explicit Jev decisions. One Bend executable owns your processes, keeps bounded logs, and asks Jev only when told. Agent skill + curl install.

Agent Skill that offloads bulk yes/no, pick-one-option and rating judgments to TypeSafe's Jev so Claude spends fewer tokens during investigations.

Splunk app: the | jev search command asks TypeSafe Jev typed questions about your events and adds calibrated probabilities, choices and scores as fields, cached in the KV store.

Many games resolved using Typesafe AI SystemOne model

Local judge first, cloud Jev second: a privacy-first request router (llama.cpp + TypeSafe Jev)

How to use Jev from Java: typed AI decisions. A runnable Jev example.