You can now ask Gemini to create Docs, Sheets, Slides, PDFs, and more directly in your chat. No more copying, pasting, or reformatting, just prompt and download.
— Sundar Pichai (@sundarpichai) April 29, 2026
Available globally for all @GeminiApp users. pic.twitter.com/VuhlvehFuU
Update 30/04/26 – 10:22 am (IST): Google has officially confirmed this rollout. Sundar Pichai just announced on X that the ability to directly generate and download files like PDFs and Sheets is now available globally for all Gemini users.
Original article published on April 29, 2026, follows:
It looks like Google has quietly rolled out an update for Gemini that allows the chatbot to generate and package downloadable files right inside the chat window.
The feature seems to have been released to users a day ago, with at least one big thread on Reddit highlighting this much-welcomed addition.
This fundamentally changes how people extract raw data from the tool. Instead of copying and pasting broken formatting text, users can now ask Gemini to output their results directly as PDFs, CSVs, plain text documents, or bundled ZIP archives.
The system even includes a new file directory built straight into the conversation view. That helps keep generated assets organized when you ask for multiple things at once.
As you can expect, users are genuinely thrilled. The main post points out how annoying it was to constantly tell the model to drop language tags just so the text would not break in the chat.
So you can bet this will be a huge time saver for anyone using the AI for coding or deep research. One user managed to get Gemini to search their drive and write a specific script directly to a zipped folder.
But there are still some hard limitations to be aware of. People in the community tested out specific formats right away. Standard Word documents appear to work fine. Transparent image files still don’t seem to be supported.
Someone in the comments suggested that, instead of asking for a raw image, have the model create raw SVG code to bypass the image generation restrictions. But as one might expect, this isn’t always ideal or even useful in a lot of cases.
Some people also noted that the feature is still pretty buggy. A few commenters mentioned the web app crashing or the model completely failing to apply specific file edits without regenerating the entire output from scratch.
Google has not officially documented the exact scope of supported file types yet. We know it handles basic structural and text formats well enough.
If you rely on Gemini for daily code generation, this update basically ends the tedious process of transferring outputs to a local editor. You just ask for the files and download the package.

