Text Summarizer
Summarize long text into a short summary with an on-device AI model. Choose short, medium, or long. Free, private, no upload. Runs on your device — nothing is uploaded, nothing is stored.
Runs a small AI summarization model entirely in your browser. The model downloads once from a CDN on first use, then summarizes text on-device — nothing is uploaded.
Paste a long piece of English text and this tool writes a shorter summary of it. Pick Short, Medium or Long, press Summarize, and the summary appears below with a word count for both your text and the result. You can copy the summary with one click.
It uses a small AI model, distilbart-cnn-6-6 (the Xenova/distilbart-cnn-6-6 build of DistilBART), which writes new sentences rather than only picking lines from your text. Short asks for a summary of roughly 15 to 60 tokens, Medium 40 to 130 and Long 90 to 230. A token is a word piece, a little shorter than a word on average, so these are not exact word counts.
Why this runs on your device
Your text is summarized on your own device: the model runs in your browser using WebAssembly, and the text you paste is never sent to a server. That makes it usable for meeting notes, drafts and other writing you would not hand to an online summarizer. The one network request is the model itself. About 284 MB of model files download from a CDN the first time you press Summarize, and your browser keeps them so later summaries start without downloading again. Until that first download finishes there is no way to use the tool offline.
- Paste or type your English text into the box.
- Choose Short, Medium or Long for the summary length.
- Press Summarize. The first time, wait for the model to download.
- Read the summary, check it against your text, and copy it.
- The first summary needs a download of about 284 MB, and the tool shows a progress bar while it loads. On a slow connection or a metered plan that is a real cost. The model also runs on your CPU through WebAssembly, so summarizing a long text takes noticeably longer than a hosted service would.
- The model was built for English news-style text. Other languages give poor or unusable results, and the tool does not translate.
- The tool does not split long input into chunks. The model reads only up to its input limit, about a thousand tokens or a little under 800 English words, so a very long document is summarized from its beginning only and the rest is silently ignored.
- A generated summary can leave out important points or state something slightly wrong, because the model writes new text instead of quoting. Treat it as a first pass and check it against the original before you rely on it.