The Short Answer
In Google Search Console, "multimodal" is a search type that counts the web searches where someone used an image instead of, or as well as, typed words: a photo taken with Google Lens, Circle to Search on Android, an image uploaded to Google Search, or Chrome's right-click "Search this image". Google announced it on 24 September 2026 and it is rolling out globally.
It tells you how people who searched with a picture found your site. Searching inside your own images, video and audio by what they contain is a separate problem, and a separate kind of tool.
What the multimodal search type counts
Google's announcement says the new data covers "searches with Lens, Circle to Search on Android, image uploads to Google Search, and the Chrome right-click 'Search this image' feature". The Search Console help page defines the filter in one line: Web: multimodal "includes web search results where an image was used as part of the search", while Web: text-based covers searches typed into the standard search bar.
So the split is by how the search started. A camera search that lands on your product page counts as multimodal; the same page reached from a typed query counts as text-based.
Where to find it in Search Console
Open the Performance report for Search results and use the search type filter. The help page lists the options as web (text-based or multimodal), image, video and news. Google says the same filter also appears in the report for Generative AI features, and that the data can be exported like any other Performance view.
You will only see figures if your site receives traffic from these searches. A site with no image-led visits shows nothing under the filter, which is an answer in itself.
What to do with it
Is this the same as multimodal search?
Only loosely. In technical writing, "multimodal search" usually describes searching a collection by more than one kind of content: finding the scene in a video where someone says a phrase, the photo that matches a sketch, or the page of a PDF that contains a chart. That is a system you build over your own files.
Google's filter is narrower. It measures Google's image-led searches arriving at your public web pages. It says nothing about whether the images, video and audio you store can be searched by what is in them.
If that second problem is the one you have, the tools are different: multimodal search APIs compared, reverse image search over your own library, and how multimodal search works end to end. Mixpeek does the second kind: it indexes the video, images, audio and documents already in your bucket so your software can search them by what they contain.
Frequently Asked Questions
What is "Web: multimodal" in Google Search Console?
A search type in the Performance report that counts web search results where an image was used as part of the search, for example a Google Lens photo or Circle to Search. Typed searches are reported separately as "Web: text-based".
Which searches count as multimodal in Search Console?
Per Google's 24 September 2026 announcement: searches with Google Lens, Circle to Search on Android, image uploads to Google Search, and Chrome's right-click "Search this image".
Why don't I see any multimodal data?
Google says the metrics appear only if your site receives traffic from these searches, and that the feature is rolling out globally from 24 September 2026.
Is Search Console's multimodal filter the same as a multimodal search API?
No. The filter reports how people searching Google with an image reached your site. A multimodal search API lets your own application search images, video, audio and documents by their content.