typesafe's jev is fun! live demo you can play with: typesafe-demo.val.run
jev-ego
A TypeScript browser agent for ego lite that turns one natural-language goal into CLICK, TYPE_TEXT, or SELECT on a numbered element table, with Jev choosing the operation and target in one request.
- Category
- Tools & Integrations
- Published by
- Community
- Author
- romaluev
- Added
- 2026-09-22
Highlights
- One TypeSafe request returns the operation and its target together; speculative target questions put two decisions in one network round trip.
- The whole loop - TypeSafe, the text helper, and CDP - runs inside one ego-browser nodejs process, with no Python process or Browser Harness hop.
- No screenshots: Jev consumes a structured element table, and visible controls, names, values, and text are read in one browser call per snapshot.
- Commands include serve, observe, act, suggest, step, run, flights, smoke, probe, and stop; serve keeps one TaskSpace in a loopback daemon.
- A first snapshot after navigation costs 3 CDP calls, and shadow roots, frames, canvas, uploads, and pop-up tabs stay outside the MVP.
Quickstart
pnpm install
pnpm build
cp .env.example .env # add TYPESAFE_API_KEY and TEXT_MODEL_API_KEY
pnpm exec jev-ego serve --url https://en.wikipedia.org/wiki/Main_Page --goal 'Find and open the article about Gödel’s incompleteness theorems.'
pnpm exec jev-ego step
pnpm exec jev-ego stopWatch out
MIT-licensed and not a general web-automation SDK; needs macOS with ego lite installed, ego-browser on PATH, Node 20+ and pnpm, plus TYPESAFE_API_KEY and a TEXT_MODEL_API_KEY for typing.
More like this
From the community
Posts from builders shipping with Jev right now.
A playable 16-judgment demo
AI multiple choice, not essay writing
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
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.
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
