2026年9月18日 發布的專案

Laya 以非自迴歸模型回答 choice、score 和 noul 問題，並由路由器選擇模型檢查點。 自架伺服器提供 POST /v1/systemone，讓既有 Jev 用戶端接入；底層執行的是 Laya 模型。

Hermes Jev Skills 提供模型路由、搜尋、記憶篩選、訊息分流與瀏覽器操作等可重用技能。 Jev 負責選擇模型、排序檢索片段等有界決策；宿主模型負責產生文字。 技能適用於 Hermes、Claude Code 與 Codex；影子模式只記錄路由決策，不切換模型。

Von 在一次前向運算中處理獨立問題，用於分類、是非判斷與有序評分。 它是獨立的 System One 模型，提供相容 TypeSafe 的 API，可在程序內或透過 HTTP 執行。 可用 Python、TypeScript SDK，或透過 von serve 提供本地模型服務。

Jev Browser Use 接手導覽、點擊、切換與捲動，文字輸入及最終驗證由 Codex 負責。 Jev 根據無障礙文字從已觀察的控制項選擇動作，由現有 Computer Use 連線執行。 需要 Node.js 22+、已連接瀏覽器的 Codex，以及相符的 TypeSafe 或 OpenRouter 憑證。

Simple Jev 使用相容的 Hugging Face 模型分類與評分，從下一個 token 的 logits 組成 JSON 答案，無需生成補全文字。 將共用上下文放入 state 或 messages，附上具型別的問題，送至已設定伺服器的 POST /v1/classifier。

OpenJev 從開放模型的機率讀取具型別答案，提供選擇、評分與是非判斷，主模型也支援圖片問題。 將 TypeSafe SDK 的基礎網址指向自建伺服器或文件中的代管端點，再以狀態與問題呼叫 system_one。

此函式庫利用表單描述 JSON 名錄，從擷取的文字辨識 IRS 表單與頁面類型。 將文字行、jevBackend 及 criteria.json 傳給 classifyPage。PDF 輔助功能另需 Poppler，後端需要 TypeSafe 金鑰。

JevRouter 從模型、代理、技能、MCP 工具及 CLI 中選擇，提供單次決策或多步能力計畫。 以 TypeSafe 或 OpenRouter 憑證使用 route、plan、SDK 或代理設定；標示的 demo 服務可離線執行。

jev-gateway 將程式代理的工具決策交給 Jev，其餘工作由原語言模型處理。 安裝 npm 套件，用對應 jev- 啟動器執行已登入的代理，首次設定 Jev 服務商。

Claude Code plugin: trim long Bash output with TypeSafe Jev before the model sees it

grep, but the pattern is a description. Filters lines by meaning with TypeSafe's Jev decision model: ~200 ms and a thousandth of a cent per line.

Fish-style zsh history autosuggestions ranked by Jev (TypeSafe)

Open-source, local-first social media research agent for Instagram, TikTok, and LinkedIn. Jev routes read-only steps; socai CLI captures cited browser evidence.

A tool calling chat bot built with Jev and no LLM.

Talk to your Mac. Local whisper.cpp + one Jev (TypeSafe) call per command + macOS automation.

High-throughput synthetic and pretraining dataset sifter for TypeSafe Jev. Rust streaming core, Parquet and JSONL I/O, typed Choice/Score/Noul judgments, speculative fan-out, 24.0 rows/sec measured.

Free Rust SEO and GEO toolkit powered by TypeSafe Jev: 58-rule audits, site crawls, AI citation checks, rank drift, CI gates, 15-tool MCP. Zero cost.

A decision cache for TypeSafe Jev-class models — memoize decisions so repeats are free, deterministic, and shareable. One 2 MB binary.

Open-source alternative to TypeSafe's Jev: a System One style model layer that gives typed, calibrated decisions from any open-weights LLM in one forward pass (HF + vLLM), with honest benchmarks

A Jev powered LinkedIn recruiting agent. Watch it browse relevant profiles, save links, and review evidence against your hiring brief.

Ruby client for decision models such as Typesafe Jev

Route HTTP requests by meaning. A semantic router for Hono powered by Jev.

Chrome extension: vibe-check your X posts with TypeSafe's Jev before you hit Post

Classify first. Read selectively. A portable agent plugin and MCP tool for batch text classification.

ACP and MCP adapter that bridges TypeSafe Jev with any LLM — computer use and typed decisions alongside Codex, Claude, Grok, and OpenCode.

Catch code issues before they catch you.

Replicating Jev with a local LLM

Data extraction for systematic reviews, quoted from the papers. Ask a trial report and its supplements your extraction form or a RoB 2, ROBINS-I, QUADAS-2 or TIDieR template; Jev points at the lines, every answer is a verbatim quote with its page, you check it and export the table. Files stay in your browser.

A spatial reference explorer for creators. Local Jev query choices, metadata highlights and source-linked collections.

A codebase scanner that helps you not run malicous code

Jev for Chrome: drives the tab you are looking at with TypeSafe Jev, a sub-second decision model. Community port of browser-use/jev-ultrafast, not affiliated with TypeSafe.

Systematic software development framework for AI coding agents upgraded with TypeSafe Jev System One typed decisions

A Document OCR Router to help route pages based on content.

Independent, evidence-based map of when TypeSafe's Jev actually holds up vs. breaks down — real API-call receipts, not a leaderboard. 中文為主的雙語 repo。

Context-pruning proxy for Claude Code and Codex: Jev judges which history is still needed, measured not claimed. POC here now, heading soon into https://github.com/compozy/compozy

Open-source macOS AI computer use and native GUI automation on Apple silicon. Jev + OmniParser CoreML + Apple Vision OCR. Bring your own OpenRouter, Vercel AI Gateway, or TypesafeAI token.

Command-line tool for TypeSafe's Jev AI model

Screens an AI agent's tool calls before they run and tool results before the agent reads them. An MCP proxy plus a hooks adapter for a client's built-in tools.

Semantic AI for pandas and Polars: classify text, analyze sentiment, and score DataFrame rows with natural-language questions and full probabilities using TypeSafe Jev.

2D autonomous car simulation in the browser, driven by TypeSafe's Jev decision model

Chrome extension that labels every post you scroll past on X with typed Jev judgments and a live cost counter

ALGAL is a programming language and VM for AI agent programs that wait for approval and leave receipts you can replay.

Claude Code Stop hook that blocks an unverified done: reads the transcript for evidence, asks Jev once, fails open on everything else

Everyday Stocks Status with Jev

Jev-powered plugins for coding agents

Claude Code plugin: Jev for rule checks, verbatim compaction, and prompt routing

用“Elon的工作”，享无福X！

Real-time drone swarm autonomy simulation using Jev for fast System 1 reflex decisions and collision avoidance, with optional System 2 reasoning for strategic guidance

The first general-purpose System One agent for Jev

Probabilistic control flow for Ruby and Rails - powered by TypeSafe's Jev