WTF is Jev by @typesafeai? Here’s the tl;dr ELI5: Think AI multiple choice, not AI essay writing. It doesn’t chat. It makes decisions your software can act on: “Spam or not?” “Which tool should this agent use?” “Does this need a human?” The exciting part: roughly 200x faster Show more
jev-mobile
Experimental Android sub-agent that gives Jev a bounded list of technically valid UI actions per step, with requirement-driven subgoals, durable SQLite tasks, confidence gates and escalation instead of coordinates.
- Category
- Tools & Integrations
- Published by
- Community
- Author
- Friedjof
- Added
- 2026-09-22
Tagscommunitypythonagentsmcpandroid
Highlights
- Jev chooses only among technically valid actions; it never generates coordinates, MCP calls or arbitrary code.
- Durable Core V1 was hardware-validated for notes and checklists, text input, crash recovery, persistence verification and foreign-editor protection.
- Tasks persist in SQLite with leases, checkpoints, cancellation and worker recovery, and mutations are journaled before execution.
- Low confidence, a small top-two margin, repeated states, loops, timeouts and step limits stop automation.
- Exposes six MCP tools over stdio or Streamable HTTP: start_task, get_task, get_task_events, cancel_task, answer_task and get_device_status.
Quickstart
bash
uv run jev-mobile worker --backend portal-adb --serial YOUR_ANDROID_SERIAL
uv run jev-mobile task start "Create a Google Keep checklist titled Shopping with Eier, Brot and Milch"Watch out
MIT-licensed. Requires Python 3.12+, uv, an Android device with USB debugging and a TypeSafe API key; payments and account changes are out of scope and broad app coverage is unsolved.
More like this
17.5kGitHub stars
Browser Use's ultrafast agent: Jev picks an operation and an element from an indexed action space in one request, and a small LLM writes text only when the action is TYPE_TEXT.
Tools & Integrations#community#python#browser
Communityjev-ultrafast
3.1kGitHub stars
An Android overlay that reads visible WeChat, QQ, and X conversations through accessibility, asks Jev for intent, danger, and the best action, then drafts three replies and fills one in without ever sending it.
Tools & Integrations#community#kotlin#android
Communityjev-chat-jarvis
2.6kGitHub stars
A cross-platform memory layer for AI coding agents: conversations from Claude Code, Codex, DeepSeek Harness, OpenClaw, and OpenCode become Markdown memories indexed in Milvus, with optional Jev reranking through the TypeSafe API.
Tools & Integrations#community#python#agents
Communitymemsearch
From the community
Posts from builders shipping with Jev right now.
AI multiple choice, not essay writing
Screening agent actions with Jev
Tested TypeSafe’s Jev (no-text, probability-only model) as an AI agent safety monitor. Checking each action first worked well caught most attacks with almost no false blocks, and much faster than Gemini.
@CompleteSkeptic
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x
Cua's small System One models
1/ Introducing CUA-S1: a family of System One Models, small, specialized, and built for computer use. Today we're open-sourcing CUA-S1-FORMS, the first in the family: github.com/trycua/cua
A 706K-parameter form filler
cua open sourced a 706k param model that fills a whole form in one 50ms pass the llm agent doing the same form took 23 turns and 39.6 seconds the specialists are going to eat the generalists from the bottom
1/ Introducing CUA-S1: a family of System One Models, small, specialized, and built for computer use. Today we're open-sourcing CUA-S1-FORMS, the first in the family: github.com/trycua/cua
Navigating Neo4j with Jev
Jev 这个 waitlist 还是很给力的，昨天申请，今天就能用上。 给已经拿到 API、但还不知道怎么玩的人整理了一份 Awesome Jev，目前我能确认到的 Jev 项目基本都在这里： 1. jev-ultrafast Browser Use 做的高速浏览器 Agent。Jev Show more
前 OpenAI 研究员 Diogo Almeida 创办的 TypeSafe AI 推出新模型 Jev。它有点像一个能读懂自然语言的超级分类器，不生成文本，只返回选项、分数和概率，专门给软件做判断。 普通大模型需要一个 token 一个 token 往外生成，Jev 则可以并行给出多个结果。TypeSafe 还用新的 RLCD
Reranking 33,047 catalog entries
拿 Jev 做搜索重排，我先泼一盆冷水：单独用，它没打赢向量检索 TypeSafe 的 Jev 这阵子很火，一堆项目拿它做重排。我们在 Agent Skills Hub 的 33,047 条目录上认真测了一次，164 条中英文真实查询，9,831 对分级标注，整套只花了 2.6 美元 三个结论 01｜单独重排，约等于没赢 Jev 重排 bge-m3 Show more
有美团、阿里的老哥嘛？ 试试加一路召回、重排（离线、近实时实现），我觉得有奇效 他在文本理解上 跟之前机器学习、llm很不一样 还能自动打标签做特征