Science, made readable.

Real research papers, rewritten in plain words: every finding kept, nothing dumbed down.

The library

Every paper already made readable. They open instantly.

Your keys

Only needed for the big commercial models. Keys stay in your browser and go straight to Anthropic or OpenAI, never to us. You pay your provider for what you use (about $0.01 to $1.20 a paper).

Keep science readable, and free

This site is free, with no ads and no accounts. The rewriting runs on a graphics card in my home, and on launch days on a rented one, so every paper someone reads has a small real cost.

If this helped you, a student or your kids understand real science, you can help keep it going.

Give on GitHub Sponsors

GitHub Sponsors takes one-time or monthly gifts, and GitHub keeps no fee on gifts from personal accounts. Organizations: sponsor a launch day or more GPU time, and your name goes on the About page. Reach me on X, @MaximeRivest.

How it works

Scientific papers are written for other scientists. This page rewrites a whole paper for a curious 14-year-old: in the authors' own voice, keeping every result, number and caveat, but in words anyone can follow.

The rewriting is done by small open models we trained (Qwen 3.5 with 9, 4 and 0.8 billion parameters). They learned from about 4,400 papers rewritten by Claude Opus and GPT-6 Astra, with plain definitions of each paper's terms looked up in Wikipedia. They run on one consumer graphics card (an RTX 3090) in my home, so a paper may wait in line, and the page may say the home GPU is offline.

How good is it? On our test papers, a judge model scored our 9B 7.8 out of 10, a little above GPT-6 Luna (7.5), below Claude Opus (8.5); it costs about half a cent per paper. It still makes mistakes more often than the big models, so check the original before relying on a detail. With your own API key you can compare with six commercial models.

Which papers? Open-access research articles in PubMed Central under the CC BY licence, which allows republishing an adapted version with credit. Figures and tables are always the original ones.

A demo, not a product. We count visits on our own server, without cookies: each visitor is an anonymous code that changes every day, and what you search for is never stored. If your browser asks not to be tracked, nothing is counted. Rewrites made by our models are kept so the next reader gets them instantly.