Computer use built with Jev

A computer-use agent built on Jev lost its first fight, then came back and won.

On every keystroke Jev picks the page's relevant tool, fills in the arguments and says how sure it is. Here it does the grocery shopping.

A backlog of hoarded X bookmarks, run through Jev: light work for a model this fast and cheap.

Clonk uses Jev to launch agents on a canvas instantly, without waiting on a slow LLM loop.

Type "the pdf I just downloaded" and the newest PDF is already the top hit, with Jev's confidence on every keystroke.

Cline's jev-browser plugin launches Chrome and lets Jev drive any browser task from inside Cline.

Jev watches the macOS clipboard; if it holds a terminal command, a quick action offers to run it.

Type a task and watch Jev pick every action from the page's action space on a Notte browser, with a step-by-step replay.

Flags AI slop in the feed in real time, as you scroll.

A browser agent where a single Jev request picks both the action and the element to click. It runs a full Google Flights search in about seven seconds.

Jev driving the computer, much faster than an LLM loop.

Jev with headless Chromium navigates Wikipedia links from one article to the other in 20 seconds.

Press ⇧⌘Space and type what you remember; Jev ranks the saved screens across apps and displays. Open source.

OpenAI's Luna builds the list and Jev handles the fast decisions to fill a real Giant Food Stores cart, cheaper than a frontier model alone.

A smarter copy and paste, built on Jev. The author thinks every computer interaction will get rewritten like this.

Say "open Notes and create…" and the app opens before you finish the sentence.

Jev with Playwright reads ~26 listings a minute, skips the misfits, bids on the good ones and messages sellers when details are missing.

ego lite with Jev and DeepSeek Flash made the same 20 calls as GPT-5.6 Sol, with the same 10/10 result, in 3.71 s instead of 54 s.

Agentcard holds the card safely, Kernel runs the browser and Jev makes the fast shopping decisions.

A macOS app applies your rules to every download: an invoice gets renamed and filed. Jev is the only model involved.

Talk to a real browser: the transcript goes to Jev, which answers in about 300 ms and clicks. $0.0002 per decision.

A local CoreML model finds the UI elements, on-device OCR reads them, and Jev picks what to click. No pixels leave the Mac.

Observe the page, send the accessibility tree and actions to Jev, let Stagehand execute. This task cost $0.001 in a remote browser.

A Chrome extension where Jev decides and agents browse, click and fill in sites right in your browser.