openread

OpenRead is a browser extension for reading web pages written in a foreign language. It translates the page in place, keeping the original layout and text visible alongside the translation, similar in spirit to the popular Immersive Translate extension. The key difference is that OpenRead does not come bundled with its own translation subscription. Instead you connect your own AI provider account, using any service compatible with the OpenAI chat completions format, and pay that provider directly for translation requests rather than paying OpenRead.

It is built as a Chrome extension using the newer Manifest V3 format, and is aimed at careful reading of articles, documentation, technical posts, GitHub project pages, and newsletter style essays where preserving the page's original structure matters. You can switch between viewing the original text, the translation only, or both side by side, and translation happens gradually as you scroll rather than all at once, so the page stays responsive. You can also right click selected text to get a translation in a small draggable popup, or translate what you type into text boxes.

For pages that need special handling, OpenRead supports what it calls Site Rules, saved instructions in a structured file format that tell the extension exactly which parts of a specific page or website should be translated, which is meant to handle tricky page layouts more reliably than a generic approach.

The project is still described as early developer-mode software, meaning some complex pages may not translate cleanly yet. To install it today you download a release file from GitHub and load it manually into Chrome's developer mode, since it is not yet available through the Chrome Web Store. To build it from source, you need Node.js and the pnpm package manager, then run pnpm install and pnpm dev. Your API key and settings are stored only in your browser's local storage and are never sent to OpenRead's own servers, only to whichever AI provider you configure. The project is licensed under Apache 2.0, a permissive license that allows free use including commercial use.