Launches on September 23, 2026

Semantic values for deterministic TypeScript: is, score and choose as ordinary values, backed by a decision model.

OpenClaw, powered by Jev. Jev-powered memory decisions with your existing LLM as fallback. Built with Jevify.

Autonomous adaptive Minecraft bot: Jev (TypeSafe System One) decides fast, an LLM plans, designs and talks. mineflayer.

Jev (TypeSafe System One) plays Doom

Goal-directed React interfaces powered by TypeSafe Jev, with a static showcase of recorded examples.

Human-in-the-loop resume and job-description matching powered by TypeSafe Jev

System One structured-answer API on DiffusionGemma, in pure Rust on AMD GPUs (CubeCL/HIP), with image input

Rock Paper Scissors against TypeSafe Jev in a terminal UI

Stop coding agents from overclaiming. Claude Code plugin + Agent Skills that verify 'tests pass' / 'fixed' / 'done' claims with calibrated TypeSafe Jev confidence scores. Works with Codex, Cursor, Gemini CLI.

Pi extension that routes extra tools through Jev so the model keeps built-ins plus need until it asks

Job-fit assessment with Jev (TypeSafe) and tailored one-page resumes, as agent skills and a Python/MCP tool

Pong in the browser against TypeSafe AI's Jev model

Kev decision models, with MiniCPM5 base support (fork of jaredpalmer/kev)

The compatibility harness for kime. The TypeSafe (Jev) API and SDKs, jev-ultrafast and Laya's own clients and tests, run against kime so that drop in replacement is a number rather than a promise.

Go samples for trying TypeSafe's Jev (System One) API with the standard library only

Lathe — Non-Generative A2UI v0.9 Compiler & Runtime (shadcn/ui × AG-UI × TypeSafe System One)

MCP server exposing the Laya System-1 decision engine (Jev) as agent tools

A Jev-compatible REST API wrapping the Laya System 1 decision model

Self-contained, ultra-fast System 1 decision engine with WebGPU/WASM acceleration, 100% wire-compatible with TypeSafe Jev (/v1/systemone).

LinkedIn feed AI-slop detector using TypeSafe Jev API

Local-first web app that pulls a GitHub repo's issues and README, then asks Jev (TypeSafe AI's decision model) to rate each issue by complexity, criticality, effort and relevance. Filter, dismiss, sort by weighted priority and export as plain text. Keys stay in memory; results stay in your browser.

A source-grounded method for producing newsletters and long-form writing with staged AI work and TypeSafe JEV quality gates

Expense tracking for Gaza households, built around bank statement exports and their note field. Uses Jev (TypeSafe System One) for typed, confidence-aware transaction categorisation.

Scans shop URLs and uses Jev (Typesafe AI) to classify items by fiber composition

Readable AI checks, measured decision policies, and inspectable reports.

Busca de restaurantes Michelin com o Jev (TypeSafe AI): 6.802 restaurantes, amostragem por rodadas e mapa mundial

Mina, an artificial person who feels time pass. TypeSafe Jev (System One) watches her body and senses every second; an LLM (System Two) decides only when a feeling fires. Nothing is scripted. Runs in Docker.

A live tone-target game scored by TypeSafe Jev.

Local MCP completion checks and repeat-loop detection for AI agents. Jev BYOK, deterministic policies, and a free offline demo.

Pluggable, observable RAG MCP server — hybrid retrieval (BM25+dense+RRF), BEIR-benchmarked rerankers (BGE / TypeSafe Jev / LLM), Streamlit dashboard + Langfuse, multimodal ingestion

From news to probabilities: a news aggregator that estimates the events listed on Polymarket, compares the estimate with the price and says what is worth doing.

Decides what to do with a failed BullMQ job. Rules answer the errors they know, TypeSafe Jev answers the rest, and the queue moves only when the answer is sure enough.

TypeSafe AI (Jev) driven, type-safe notification routing engine for Go.

One box for Omarchy: apps, menu actions, inline maths and conversions, search shortcuts, URLs, and optional TypeSafe Jev matching by meaning

MCP plugin to connect jev-like system one API (local hosted or typesafe jev) to codex and claude code for decision tasks like compaction, verification judgement, etc.

Linux-only OpenCode shell permission guard using Jev; auto-approves simple reads and limited /tmp writes.

pairsort: rank anything with AI judges — many small pairwise questions, coupled into one calibrated ranking (PKPD / Bradley–Terry) on Jev-style judges

better context compaction with TypeSafe Jev judging what to keep

tool result filter with TypeSafe Jev to keep main agent context smaller

Route mid-run Pi prompts with TypeSafe Jev

Adaptive reasoning-effort router extension for Pi using TypeSafe Jev

One arcade: Heist, Chess, and Minesweeper with TypeSafe Jev through Vercel AI Gateway.

R³∞ strato riflesso System One — candidato CA. Decide, non scrive. Non è canone.

Ultra-fast, sub-100ms API & webhook guardrail and triage gateway powered by TypeSafe AI System One (Jev). Parallel 7-dimension speculative evaluation, deterministic policy router, zero-dependency SQLite audit trail, and automated outbound dispatch.

A daily survival writing game judged by Jev.

An ant colony where every ant asks Game: TypeSafe Jev what to do next — you are the queen, and your only control is a text field.

S1Rank: Can a System-One decision model (TypeSafe Jev) rerank? Benchmark, raw responses, and paper.

Give Claude Code a second opinion it can ask for in half a second. Reusable "battery" checks plus optional risk and finish gates, powered by TypeSafe's Jev.

Open self-hosted System One decision API (Jev wire-compatible) + agent skill — 87.9% JevBench, beats Jev on accuracy and score

Jev-powered model router for AI coding agents (omp + Claude Code): small jobs go to smaller, cheaper models.