Launches on September 17, 2026

Kev provides pretrained decision models and fine-tuning for yes/no, multiple-choice and rating questions over shared text. Its API follows TypeSafe System One, allowing the TypeSafe Python SDK to target a local Kev server.

This npm library and Claude Code plugin compact history by pruning tool calls and results instead of rewriting conversation text. Jev answers separate noul questions about retaining each call and result; thresholds determine whether to keep, shorten or remove the pair.

The model answers choice, Boolean and ordered-score questions from states and candidates without decoding answer tokens. NanoJev is a separate local model inspired by Jev; recorded comparisons include Jev and untuned Qwen on the same game interfaces.

TypeLLM adds JSON Schema-constrained outputs to existing autoregressive models without changing their weights or architecture. Inspired by Jev, it retains free text and optional reasoning and supports dependencies between output fields. Serve a compatible model with SGLang, then submit context and typed questions through TypeLLMClient.

Foreman supervises coding workers, assessing completion, requirements, test sufficiency and the need for human input. Jev evaluates checks; a deterministic Python arbiter selects whether to continue, steer, verify, stop or escalate. Supply a repository and job with Python 3.11+, a TypeSafe key and a configured coding-agent backend.

Jev Search turns plain-language requests into web links and snippets with editable filters and visible relevance scores. For local use, configure Search1API and a Jev provider in .dev.vars, install dependencies with pnpm, and run pnpm dev.

Jev MCP gives agents tools for evidence verification, classification, reranking, extraction and patch review. Register npx -y @jkudish/jev-mcp in an MCP client using Node.js 22 or newer and a configured Jev provider key.

Mobile Jev controls an Android device through Mobilerun, with a live studio, CLI and execution traces. Configure both provider keys and a ready device ID, then use pnpm dev to open the local studio and submit a goal.

pg-jev filters, ranks and classifies PostgreSQL rows using plain-language SQL conditions. Batched Jev questions judge rows; SQL functions expose boolean, probability, choice and rubric-score results. Requires PostgreSQL 14–17, plpython3u, superuser access and a TypeSafe key.

jev-voice-browser streams speech transcripts to a Node server controlling a visible Chromium window through Playwright. Jev selects intent and targets; code extracts literal text candidates and decides whether to act, wait or ask. Use Node 20+, a TypeSafe key and Chrome or Edge for the microphone control page.

System One Connector exposes typed decisions to coding agents through its evaluate command and MCP integration. It connects to Jev or compatible self-hosted models such as CLM and Laya. Define Choice, Noul or Score questions; agent setup and batch evaluation are provided.

Unclutter hides nonessential elements in Chrome, Chromium and Firefox using saved rules for matching page templates. Jev classifies candidate elements into typed categories; extension code validates responses and controls selectors and reversible hiding. Choose TypeSafe or Vercel credentials and analyze manually by default. Saved rules continue working offline.

The project pairs Qwen3.6-35B-A3B on SGLang with a separate Python API process for typed decision serving. POST /v1/systemone supports Noul, Choice and Score; jev-latest is an alias for the configured Qwen model, not a TypeSafe request.

This learning project answers choice, yes/no and ordered-score questions about one image by directly scoring candidate outputs. It is an independent community exploration of Jev-like inference, without reproducing TypeSafe's proprietary architecture, training, calibration or serving system.

Jev Browser runs a task from a starting URL and returns the final page, screenshot and decision trace. Use the npm CLI, MCP server or navigate library with Node.js 22 and a Jev provider key.

Self-hosted, versioned skills library for AI agents. MCP, scoped clients, and optional Jev recommendations.

Describe your startup idea. Jev decides: kill it, fix it or ship it.

Local-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev.

"Work appears completed or recorded. Run /compact to save tokens."

Live Jev trader on Hyperliquid

A small open decision model: state + typed questions -> calibrated probabilities. A Jev / System One re-creation on Qwen3.5.

A skill for writing and improving programs that call Jev, TypeSafe's System One model

Rust CLI powered by Jev from TypeSafe.ai that ranks agent skills for the next step using live session context. Includes Claude Code hooks, structured JSON, abstention, and local feedback. Requires a TypeSafe API key.

Open Jev implementation with custom finetuning

Bounded TypeSafe Jev workflows for coding agents.

Non-autoregressive decision engine on ModernBERT (151M) with calibrated uncertainty (RLCD), TypeSafe AI Jev benchmark audit, and in-browser WebGPU playground

A fast browser agent: Jev picks each action from what is on the page, an LLM reads and plans, and every claim in an answer cites a quote from the page.

Personal-assistant agent built on Vercel's eve with 100 mocked tools, measuring how many steps it takes when Jev picks the tool versus the LLM.

Put a live Jev (TypeSafe) meter on any video: every sentence scored, rendered as a 16:9 edit

Using Jev as an evaluator.

CLI that picks Cursor, Claude Code, Codex, or OpenCode + model/effort for a task, then launches it. Powered by Jev and Herdr

Pre-alpha PostgreSQL extension for TypeSafe AI (Jev) categorical classification

🧹 Fun project: a Chrome extension that asks a tiny AI decision model (TypeSafe Jev) "is this DOM element an ad?" and pops it off the page. BYOK, no backend, not a real ad blocker.

Ask your house a question, get a number back. Home Assistant integration for TypeSafe Jev: typed answers as sensors, four actions for automations, and a conversation agent for Assist.

MCP server that puts TypeSafe Jev on the coding loop in Cursor, Codex, and any MCP client

mini-Jev: what a Jev-style typed-decision interface looks like on a frozen Qwen3-4B — read the option letter's logits instead of generating JSON. Preregistered experiment, results, teaching bench.

OpenDecision is an open-source semantic decision engine like typesafe's jev.

Semantic tool routing and typed System One decisions for the Pi coding agent using TypeSafe Jev

Real-time Discord moderation bot: Jev evaluates messages and metadata in parallel to catch phishing, spam, and social engineering with a progressive escalation ladder.

Turn any off-the-shelf LLM into a Jev -like decision layer

Auto mode for every coding agent, built on Jev: risk-scores every tool call with session context (deny / ask / allow), flags prompt injection in results, checks skills and plugins. Claude Code, Codex, Copilot, Gemini, Cursor, pi, OpenCode, ACP.

Skill for Hermes, and other agents, to ask typesafe's jev

1v1 Jev quickscope arena — Three.js + TypeSafe System One

🚀🚀 A 0.8B JEV-like multimodal model playing GUI games directly from raw pixels.

Open replica of TypeSafe's Jev: typed calibrated decisions in one forward pass, on Gemma 4 E2B / Gemma 3 270M (Modal)

Classify Git commit diffs and messages with Jev. Bug fixes, security fixes/CWEs, and change types.

An agent skill to discover TypeSafe Jev opportunities, design typed questions, and learn from recent community experiments.

Trading bot with the all new TypeSafe AI's first system one model named as Jev

Batched single-token choice inference for open language models, compatible with TypeSafe

TypeSafe Jev for OTP: reply to Jev from a GenServer and pattern match on its answer