GoogleapiOffice Files and Multimodal Tool Results in the Gemini API
Small developer update: the Gemini API now reads more Office and OpenDocument formats, and tools can hand files, audio, and video back to the model.
Details
- Accepts PPT, ODP, and ODS files as input, and adds page-image processing for older DOC files
- Tools the model calls can now return Office and OpenDocument files, audio, and video, including files already uploaded through the Files API
- The same file support applies to tool results in Google's Interactions API
Best for
Building agents that work over slide decks, spreadsheets, and older Word files.
When to use it
Your app feeds PowerPoint or OpenDocument files to Gemini, or your tools return documents and media rather than plain text.
Who benefits
Developers building document and file workflows on the Gemini API.
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →