What's newin version 8.25
New Text to Speech Voices — We are rolling out new (much more lifelike) voices in beta! They will eventually be updated to be the default voices, but in the mean time you can choose your favorite from the TTS voice selection sheet.
Reading Themes — You can now choose from a variety of different reading themes, including Sepia, Slate, Soft, Ink, and more.
GPT-6 Luna For Ghostreader Summaries — Tristan and Ibai upgraded Ghostreader’s default model for preset prompts, including the automatic summary and document tagging, so they’re now powered by GPT-6 Luna. As with previous default models, this is included with your Reader subscription.
YouTube Chapters — Newly saved YouTube videos now show their chapters in the Contents panel, so you can jump straight to any section.
Faster Heavily Highlighted Books — Books with lots of highlights now open much faster.
Dozens and dozens of new bug fixes — PDF dark mode no longer resets when you open a sidebar, large PDFs download reliably, EPUB tables of contents stay correct when you switch books, bracketed tag names no longer crash the app, "Save to" no longer crashes with long book lists, highlights made across links and line breaks anchor correctly, swiping through your feed no longer skips articles, text-to-speech handles plain-text feed items better, tags and text scale with larger system font sizes, and Ghostreader prompts now apply their changes properly.... and so much more.
Previously:
Mobile Chat — The long awaited “chat with document” feature is now available on mobile. Intuitively ask any question while reading and receive a fast, accurate answer that will link to the cited text in the document. Ghostreader is also aware of your position and context while reading -- it’s like a smart friend reading over your shoulder that you can ask questions. To try it, just trigger Ghostreader while reading.
Modern Toolbar — We’re also introducing a new mobile toolbar in the reading view! You can enable it in the app from your settings at Account > “Modern Toolbar”. The new toolbar allows you to quickly access ghostreader (in one tap), gives you more easy access to delete documents, and allows you to jump to the previous/next documents without needing to leave the reading view. Of course, if you prefer the existing toolbar, we have preserved that option as well.
Upload Markdown — You can now upload markdown documents to Reader.
Better Search (beta) — There is now a toggle on the Search screen to enable our new server-based search that runs both more quickly, accurately, consistently, and takes much less time to index all of your documents (compared with our old local-first search).
Lookup in Reader — Reader now has a dedicated way to look things up without leaving the page. Easily select (long press) any word or phrase and Reader explains it right away in a dedicated lookup panel. For stand words/terms, get standard definitions, or when you select a term (or name) unique to the document, get the definition the way the author uses it. If you want to revert to the old selection behavior, there is a new option to do that too.
Readwise MCP & CLI — The (new, improved) Readwise MCP server gives Claude, ChatGPT, Cursor, or any MCP-compatible AI direct access to your Readwise highlights, and the full text of all your Reader documents. Once connected, you can ask your AI to search your notes, answer questions using your own reading history as context, or even make changes to your Reader library for you. We also pre-built AI workflows like inbox triage, feed catchup, a custom book review that is aware of related context in your vault, and a quiz mode for testing yourself on recent reads.
Podcast Transcripts — share any podcast episode from Apple Podcasts, Spotify, or Overcast into Reader, and it will be saved as a permanent, highlightable transcript.