JevMade field notes / TypeScript tutorial

Categorize text inside a web browser using open-jev

The open-jev project lets web browsers categorize text locally. This guide shows how to load these AI decision tools, ask multiple-choice questions, and manage device limits without sharing private data with an outside server.

Original by Nico MartinIntegrations

Listen to this guide

JevMade’s plain-English explanation

0:00 /

AI narration

Credits and license

“Run typed decisions locally in the browser with open-jev” by Nico Martin. Read the original source.

This expanded guide is an AI-narrated adaptation of the source’s essential explanation, examples and caveats, not a word-for-word reading. The synthetic voice does not imitate the author or imply their endorsement.

License: MIT

MIT License

Copyright (c) 2026 Nico Martin

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Our summary

The open-jev project is a tool for web browsers that categorizes text. It copies the interface of Jev, an AI tool that chooses from options you provide instead of writing new text. People use this to sort incoming messages without sending private information to an outside server.

The software downloads a small AI tool, called a model, directly to the user's device. When given a piece of text and a list of questions, the software calculates the probability for each possible answer. It uses the device's graphics hardware to run these calculations quickly, falling back to standard processing if needed.

Running this software locally requires a modern browser and uses the device's memory and internet data. It can only read a limited amount of text at once, ranging from a few paragraphs to several pages. This is useful for developers building private sorting tools who accept device performance limits.

Key takeaways

Check if the user's device supports fast graphics processing before loading the software.
Account for the time and internet data needed to download the required files to the browser.
Keep the text short enough to fit within the specific tool's reading limit.

More guides →See the experiments →