Table of Contents
Gemini can now create, preview, and edit files without leaving the chat
Since April, Gemini could turn a prompt into a finished Google Doc or spreadsheet. Fixing a typo in that file meant leaving the chat, opening the file somewhere else, and editing it by hand. Google closed that gap on October 29, 2026: you can now generate, preview, and edit documents, spreadsheets, slide decks, and PDFs right inside the Gemini app chat.
The feature itself is free, no new plan required. But comments, version history, and real sharing, the stuff that makes Docs, Sheets, and Slides worth using at all, still live only in the full apps, not in the chat window.
What's actually new
The April 2026 launch, "Move from conversation to creation with file generation in Gemini," only went one direction: prompt in, finished file out. You could ask for a project plan as an .xlsx or a course syllabus as a .docx, and Gemini handed back a downloadable file. Editing meant opening it somewhere else. Gemini also capped you at one file per prompt, a limit the October post doesn't mention changing.
The new update adds the loop back in. Per Google's announcement, you can generate a document, spreadsheet, slide deck, PDF, Markdown file, plain-text file, or a block of code, and the result previews directly in the chat. From there you can highlight any text to ask Gemini a question about it, expand the preview to full screen and type edits directly, or just ask Gemini to refine a section in plain language. PDF preview and editing works on both web and mobile; most of the other formats, including code, are web only for now.
Google's own framing in the post leans toward drafting, not finishing. The listed use cases are generating a presentation outline and previewing the layout "before opening in Google Slides," and building a structured tracker in Sheets to "open it directly." That phrasing matters more than it sounds. The chat preview is pitched as the step before the real app, not a replacement for it.
Who gets it and what it costs
| Requirement | What qualifies | Source |
|---|---|---|
| Account type | A personal Google account, a work account on a qualifying Workspace edition, or a school account with Gemini enabled | Google Workspace Updates, Oct 29, 2026 |
| Separate cost for the feature | None. It ships to everyone already signed into the Gemini app, free or paid | Google Workspace Updates, Oct 29, 2026 |
| Admin control | None exists. Admins who want Gemini off entirely toggle the whole app, not this feature | Google Workspace Updates, Oct 29, 2026 |
| Google AI Plus (consumer, optional) | $4.99/mo, 2x Gemini usage limits, 400 GB storage | Propakistani, Oct 17, 2026, fetched 2026-10-30 |
| Google AI Pro (consumer, optional) | $19.99/mo, 4x usage limits, 5 TB storage, Pro model access | Propakistani, Oct 17, 2026, fetched 2026-10-30 |
| Google AI Ultra (consumer, optional) | $99.99/mo standard tier, 5x usage limits | Engadget, May 19, 2026 |
Nobody pays extra for in-chat file editing itself. The paid tiers buy you higher usage caps and access to Google's stronger models generally, which matters if you're generating files all day and hit a rate limit, not because the feature has its own price tag. Workspace customers get it through their existing seat.
One small, telling gap: as of today, Google's own help article for generating files from Gemini chats still only describes the generate-and-download flow. It lists "Workspace files (Docs & Sheets)" as a supported format, no Slides, and says nothing about editing in place. Both the April and October blog posts list Slides and the new edit controls. The support documentation hasn't caught up to the announcement yet, which is a small reminder to trust the dated blog post over the help center when they disagree.
How to turn it on
Nothing to configure first. For end users:
- Open gemini.google.com or the mobile app, and sign in.
- In the prompt box, ask for a file by name: "write a one-page project brief as a doc" or "build a budget tracker as a sheet."
- When the preview renders inline, read it in the chat. Nothing has downloaded or saved anywhere yet.
- To fix something small, highlight the text you want to change and ask Gemini about it directly, or type a follow-up like "make the timeline section shorter."
- For a bigger pass, expand the preview to full screen and edit the text yourself, the way you'd edit in a plain text box.
- When it's close enough, either download the file as-is, or open it in Docs, Sheets, or Slides to keep working with other people.
Skip that step and the file has no revision history, and nobody else can touch it.
Where it still falls short
- No comments. The chat preview has no margin notes, no @mentions, no resolve-this-thread workflow. That's a Docs, Sheets, or Slides feature once you open the file there.
- No version history. Every edit inside the chat preview overwrites the last state. There's no "see earlier version" panel the way there is in a real Google Doc.
- No native sharing. You can't drop a link to the in-chat preview into Slack for a colleague to edit alongside you. Sharing starts after you open the file in the dedicated app.
- Most formats are web only. Code previews, full document and spreadsheet editing, and slide deck editing currently work on the Gemini web app; PDF is the one format that also works on mobile, per Google's own breakdown.
- Still one file per request, per the April post, and the October update doesn't say that changed.
- No admin control, for better or worse. An IT team that wants to restrict this has to restrict the Gemini app as a whole.
That makes it a drafting tool, not a Docs replacement.
The alternatives
ChatGPT's Canvas does a version of the same thing: a side panel where you draft and edit text or code with the model, with the model proposing targeted edits instead of rewriting the whole block. It's built for iterating on a single document in the chat itself, and getting that content into a Google Doc or Office file still means exporting or copy-pasting it out.
Microsoft 365 Copilot skips the chat-to-file round trip entirely by living inside Word, Excel, and PowerPoint from the start. You're editing the real file the whole time, with comments, version history, and sharing already attached, because you never left the app. The tradeoff is you need a Microsoft 365 Copilot license and you're drafting inside Office's interface, not a chat window.
Claude's Artifacts render documents, code, and simple visuals in a side panel next to the conversation, similar in spirit to Gemini's new preview. Claude doesn't generate native .docx, .xlsx, or Google Workspace files directly from an Artifact; getting the content into Office or Workspace format is a manual export step.
The verdict
If you're drafting something quick and disposable, a meeting summary, a first-pass outline, a one-off tracker, the in-chat preview skips a step that used to be mandatory. Ask, read it in place, fix the one line that's wrong, download it, done.
If the file needs a second person on it, a shared budget, a deck three people are editing before Friday, open it in Docs, Sheets, or Slides the moment you're past the first draft. Don't live in the chat preview for anything that needs comments or a revision trail; it doesn't have either.
If you're already inside Microsoft 365 and paying for Copilot there, you're not missing much by skipping this. You already get live editing with full collaboration built in, which is the piece Gemini's chat preview still doesn't have.