Summary

  • Gemini’s current image analysis is limited to single uploads. While it offers features like image description and text extraction, users must send images one-by-one, unlike ChatGPT’s multi-image capability.
  • Google might be developing multi-image upload functionality. Leaks suggest Gemini may soon allow users to upload up to 10 images in a single prompt for comprehensive analysis.
  • This feature is currently hidden in the Google app beta. Though functional, it requires manual activation and is likely to roll out widely with a future update.

Gemini, on both the web and its mobile app iterations, can analyze images to help you understand them better, offering functionality like image description generation, text extraction, object identification, and more. Unlike ChatGPT‘s implementation, though, Gemini can only analyze a single image at once.

For users that are looking to have multiple images analyzed, this limitation means that they must send images one-at-a-time — in multiple single image prompts. However, this could soon change, with Google reportedly working on ChatGPT-like multi-image uploads.

This comes soon after Google was spotted working on video analysis support for Gemini.

Google Gemini


Related


Google Gemini is working on this long-overdue feature

Support for video uploads could be on the way



1

Gemini currently throws a “replace current image” prompt message when you try to add a second image to your query, suggesting that “adding a new image will remove the current one.” You do have the option to create a larger collage of multiple photos to send in one prompt, but that ends up taking more time than simply sending images one-by-one.

With a future update, as spotted by credible leaker AssembleDebug in a report for Android Authority, Gemini could allow users to upload up to 10 images in one prompt.

Multi-image analysis could be coming soon

Screenshots highlighting Gemini's current and upcoming image analysis capabilities.

Source: Android Authority

As seen in the screenshots above, Gemini was able to analyze and share information about all images within the prompt, with a separate breakdown of each image. Users then have the option to pose counter-questions.

Although the feature’s implementation looks polished, in UI terms and in functionality terms, it still isn’t available to try out. The functionality was found hidden in Google app version 16.11.32 beta, and had to be manually enabled. Simply installing the latest beta build isn’t likely to surface the functionality for you.

Once officially available, the functionality is likely to make its way to the web first, before arriving on Gemini’s mobile app(s).