Google brings rich document editing to Gemini on mobile and web

At this point, we’ve all seen how easy it is for AI to generate some documents. But there’s still often a big gap between getting that first draft produced, and fine-tuning the model’s output to the point where every last element is just how we want it to be. Google’s always working on tweaks to improve that process, and today the company shares some of its latest progress.

Over on its news blog, Google is highlighting some of the new ways Gemini is learning to work with files like PDFs, spreadsheets, presentations, and even code. Users can now easily explain their edits to Gemini and get previews straight away, combining text and visual input to refine what Gemini creates.

If some of this sounds a little familiar, we don’t blame you — Gemini is just so multi-faceted at this point, and shows up in so many different places, that it’s hard to avoid a little overlap. And sure enough, Gemini’s been creating and editing documents for us already. For this, though, Google’s focusing explicitly on being able to pull all this off right from within the Gemini app itself, complete with rich, interactive previews.

Right now, Gemini on the web enjoys the broadest support across different document types. That includes everything from Workspace cloud files, to even stuff like Markdown. If you’d rather use Gemini on your phone, the app’s limited to working with PDFs at the moment.

That restriction aside, access here is pretty wide open, and Gemini users everywhere should be able to get started creating and revising documents in all supported languages. Give it a try for yourself, and let us know how well Gemini performed — or tell us where Google itself needs to make a few edits — down in the comments.