Launches on September 20, 2026

A simulated drone flown through an obstacle course by Jev, TypeSafe's System One decision model

Free English RAG benchmark for TypeSafe Jev 1.13 — frozen candidate pools, calibration, paired bootstrap CIs, $0 runs. Compared with OpenJev and NVIDIA cross-encoders.

Code review filter for AI Agents using TypeSafe Jev model

System One\Two Driven Robot demo

Jev vs GPT-4.1 as synthetic survey respondents on Twin-2K-500. How you ask mattered more than which model you used.

Table tennis vs. TypeSafe's Jev (System One). Every paddle move on the right is a live model decision — no local prediction, just a lookup table and a servo.

Hermes plugin: drive and read a visible terminal-browser tab with Jev typed decisions. No screenshots, fractions of a cent per step.

Watch TypeSafe Jev play Slay the Spire 2, with a local decision dashboard and offline evaluations.

The doorman for Claude’s context window. JEV-powered model routing and recoverable context filtering for Claude Code.

Voice-controlled browser via Jev, native CDP (port 9229), Doubao Voice Bridge and Web Audio streaming

Zero-shot benchmark of TypeSafe Jev 1.13 (decision model) vs open-weight non-generative models PrismNLI-0.4B and Laya: frozen protocol, raw predictions, calibration, latency, report

Java SDK for the TypeSafe AI API. Typed questions, typed answers, Java 21.

Jev-driven interfaces with on-demand diffusion visuals and constrained rendering.

JEV core — the judgement primitive and harness for TypeSafe System One (Jev). Closed-set, fail-open, stdlib-only.

Turn Any Open-LLM into a System-one Jev model

jevkit for JavaScript/TypeScript: static linter and shared record format for building on TypeSafe's Jev (System One) model. Unofficial.

jevkit for Python: static linter and shared record format for building on TypeSafe's Jev (System One) model. Unofficial.

An attemt to recreate jev model on top of gemma

1,000 synthetic SaaS accounts, 6,000 constrained judgments from TypeSafe's Jev, and ordinary TypeScript deciding who needs a human today. TanStack Start on Bun.

GitHub Action + CLI: triage PRs with TypeSafe Jev (ready / needs review / risky) with confidence gates and jevcheck-friendly contracts.

Generate a landing page from one sentence: an LLM writes the copy, Jev makes the judgement calls, code owns the rules.

Typed AI decisions for Codex, Claude and any MCP client, powered by TypeSafe Jev. Classify, score and evaluate with one generic tool.

Minesweeper where deterministic logic does the provable work and TypeSafe Jev is consulted only when the board forces a guess.

Verbatim context compaction for OpenClaw: a context engine powered by TypeSafe's Jev. Drops stale tool calls and results, never summarizes.

Effect-native OpenCode coordinator with Jev routing and persistent goals

Jev-governed AI agent harness: every step routed through TypeSafe Jev decisions

Open-weight System One model (Jev-compatible): calibrated noul / choice / score decisions in one forward pass, no text generation. Works with TypeSafe's typesafe-sdk unchanged.

A Jev (TypeSafe System One) permission gate for the Pi coding agent: judges every bash / write / edit call before it runs

👩‍🔬 Pilot of the decision model 'jev' from TypeSafe AI

Cloudflare Worker to check for profanity using TypeSafe Jev

TUI harness orchestration with herdr and jev to efficiently distribute work across your coding subscriptions.

Native System One inference gateway for HTTP and MCP.

Skill Dash uses Jev to judge Claude Code and Codex skills and plugins: usefulness, redundancy, clarity, duplicates, safety. Local dashboard plus the corpus pipeline behind whichskills.dev. MIT.

Mop the slop out of your LinkedIn feed. Not an AI detector: a bad-writing detector. Chrome extension + Vercel server, powered by Jev from Typesafe AI. MIT.

Claude reads the part of a large file that answers your question, not the whole file. A PreToolUse hook. Measured: -36% cost on the reads it fires on, no answer lost.

Readable scripting, semantic code checks, a Go System One client, and Studio.

A universal, sub-100ms FOSS gaming bot framework powered by OpenCV for lightning-fast visual telemetry and Jev System 1 AI for non-autoregressive decision loops.

Windows port of typesafe-computer-use with Jev decisions, local OCR/UIA, and MCP host integration

Ruby client for Jev, TypeSafe's System One model: typed questions, probabilistic answers. Zero runtime dependencies.

Unofficial Ruby SDK for the TypeSafe AI API (Jev model) - typed questions, retries, and typed errors. Community port of typesafe-sdk-js.

Portable agent skill: TypeSafe Jev as a cheap code-review classifier (HTTP + optional jev-review MCP)

You say it. Your app does it. Voice and typed commands for Rails.

Independent, source-linked research on TypeSafe AI's Jev (System One), with 947 rubric-scored public repositories, recurring patterns, datasets, and bilingual documentation.

Federated knowledge retrieval engine for AI agents. Routes queries across specialized backends (FTS5, ChromaDB, graph) via a registry-first orchestrator with TypeSafe/Jev AI routing and evidence re-ranking. Read-only by default, security-gated by design.

Lightweight, zero-dependency AI Agent runtime firewall (L7 HTTP Proxy + MCP Sandbox) in Go.

AI Music Video Director - Beat-synced full-motion MV generation pipeline with TypeSafe JEV SystemOne, MiniMax Hailuo, and ffmpeg

Evidence-first internal-link auditing for SEO. Deterministic checks plus optional TypeSafe Jev judgments, exact source context, and CI regression gates.

Welke arXiv-papers helpen een HR-/People-lead bij AI-gebruik in de organisatie? Geclassificeerd met Jev (TypeSafe).

Confidence-gated action router for TypeSafe Jev — per-action thresholds, human floors, probability + confidence. Liquid gold Pattern 2.

Jev scores whether a tweet goes viral; OpenRouter writes drafts and a write-judge-revise loop climbs the score.