Launches on September 22, 2026

AI-native Android UI test runner using TypeSafe/Jev + Appium

Fast, low-cost real-time agents: take small decisions off the critical path with a System One classifier (Jev). Live demo in your browser.

A small Convex component that uses Jev to build a semantic index over your tables.

Measure an agent or a prompt with a panel of typed questions, judged by Jev

Let Jev judge. Let your code decide. Python issue-routing sandbox with labeled replay, an offline demo, and Kev support.

Typed Rust for TypeSafe AI's Jev: define questions as enums with macros, ask them with a built-in client, get typed answers with confidence.

Shared Python infrastructure for the JevKit tools

Plugin for Claude Code and Codex: TypeSafe's fast classifier Jev screens spec drift, failed CI runs and code-rule violations in seconds; your agent investigates only what it flags. Specs in English, Latin-alphabet languages, Chinese, Japanese and Korean.

Open-source MCP server that lets AI coding agents offload yes/no, multiple-choice and scoring questions to TypeSafe's Jev. Jev reads your files, your agent gets back tiny, confidence-scored answers, fewer tokens, faster runs. Works with a TypeSafe or OpenRouter key.

Jev (TypeSafe System One) に判断を委譲する関数モックデコレータ

Local daemon serving the pinned TypeSafe Jev wire surface, backed by a configurable OpenAI-compatible model.

A jev client for the live typesafe.ai api, written for a racket language runtime

A test kitchen for questions code can't answer, exploring structured AI decisions with Jev.

Ask Jev whether given curl | bash install is safe or not

An endless piano, imagined by JEV. Debussy-inspired music that unfolds as you listen.

The if statement, finally powered by AI. Every branch is decided by TypeSafe's Jev model.

The if statement, finally powered by AI. Every branch is decided by TypeSafe's Jev model.

JimmyJenk: a local harness for Jev-style System-One decisions (Noul / Choice / Score) from any LLM's token logprobs. vLLM + llama.cpp backends, playground, calibration and robustness evals.

Jev picks Claude Code's reasoning effort before every step. Low for routine, high when recovering.

Dual-Process AI Agent combining Gemini 3.5 Flash-Lite (System 2) and ConvAI Laya / TypeSafe Jev (System 1)

Head-to-head evaluation of Laya (open weights) and TypeSafe Jev on email intent classification

An HTTP server for Laya, so apps that aren't Python can use it.

Run Laya behind the TypeSafe System One API: a local, drop-in endpoint for Jev clients. Docker (CPU, CUDA, ROCm) or native Windows on a Radeon.

A town you can fork — a Dwarf Fortress–style life simulator where you rewrite any decision and watch history diverge.

Self-hosted System One decision engine (TypeSafe Jev's Choice/Score/Noul contract) running locally on LM Studio, with real probabilities from token logprobs.

Fan-made poker game set in the world of CryptoNinja's 月蝕綺譚 -Luna Occulta-, built on jev-poker (CPUs think with TypeSafe Jev). 非公式二次創作。

Code and raw data behind Lunar Terminal: Robocode Tank Royale bots, GapFlap, a token-metering proxy, and the 327-decision randomised trial of TypeSafe Jev.

Adaptive reasoning effort for Claude Opus 5.5 in Oh My Pi, decided per generation by a TypeSafe System One model.

Semantic classification of Outlook, Gmail and IMAP mailboxes with Jev (TypeSafe) — one category per message, preview by default, cost-capped, local. C# / .NET 10, MIT.

A tool-call gate for coding agents, powered by TypeSafe Jev

Minimal MCP wrapper for TypeSafe Jev. Structured yes/no, choice, and score decisions. Docker-ready. Unofficial project.

Test what an AI output means. Natural-language assertions powered by Jev, with pass, fail, and inconclusive results.

See the consequences hidden in a code diff. Jev-powered semantic reviews, a CLI, and an interactive demo.

Benchmark de triagem de revisões sistemáticas: JEV (TypeSafe) × DeepSeek V4 Flash × Qwen3-Reranker, contra o julgamento documentado de autores reais (SYNERGY+ v3.0)

BBB Market YEE-29 deterministic analytical feature layer

Minesweeper experiments comparing Jev, LLMs, hybrid policies, and deterministic solvers, with documented methods and recorded results.

n8n community node that turns Jev by TypeSafe into ready made decisions

n8n community node for TypeSafe AI's Jev System One decision model

Private, browser-only analytics dashboard for your Netflix export. 12 tabs, local parsing, opt-in AI.

typesafe jev openclaw plugins

OpenJev fork: System One decisions from an open model, plus a Jev vs LLMs explainer on GitHub Pages

Local typed decisions on ARM: a TypeSafe-compatible System One API for Raspberry Pi. llama.cpp Q8 backbone + torch decision head; 4.8x faster than the official SDK, decisions identical.

let Jev plan a SQL

Typed Jev decisions inside PostgreSQL

An advisor called Jeff for pi: calibrated decisions via ask_jeff, backed by the Jev-compatible System One API (OpenRouter / TypeSafe / jev-agent)

Ultra-lean supervisor for Pi: zero-token regex guardrails, test verification tracking, and TypeSafe Jev semantic slop gate.

Autonomous Pokemon Showdown battle agent powered by Jev AI (System One Model by TypeSafe AI), with a validated poke-env harness, safe fallback decisions, and a live dashboard for transparent battle state, legal moves, probabilities, telemetry, and Jev decision history.

Examples of TypeSafe Jev in Pydantic AI

Non-intrusive QQ chat intent & risk HUD powered by TypeSafe Jev (System One) - 无侵入的 QQ 聊天意图/风险悬浮卡

Ejemplo didáctico: captura los paquetes de tu red con Wireshark/tshark y deja que Jev (TypeSafe) los juzgue desde la ciberseguridad