> ## Documentation Index
> Fetch the complete documentation index at: https://fabric.so/llms.txt
> Use this file to discover all available pages before exploring further.

# Smart search

> Your personal search engine.

<img style={{ borderRadius:"0.5rem" }} src="https://mintlify-assets.b-cdn.net/bigbend.jpg" className="hidden dark:block" />

<Tip>
  **Please note:**

  Smart Search is only available on paid plans. [Click here to upgrade](https://fabric.so/pricing).
</Tip>

<img style={{ borderRadius:"0.5rem" }} src="https://mintcdn.com/fabric-8a249960/7jRp3Qwzy0kFFSpt/images/search-social-image-2-c.png?fit=max&auto=format&n=7jRp3Qwzy0kFFSpt&q=85&s=7abeed024cdb655ad0dbd29cd961ac63" className="dark:hidden" width="1800" height="945" data-path="images/search-social-image-2-c.png" />

## Search across all your apps and clouds

If you've connected any external data connections (Google Drive, Dropbox, Gmail and more), then making a search in Fabric will also try to fetch results from these sources.

* If a data connection is 'search-only', then the search will be keyword-based, using the data source's API.
* If a data connection is a 'connected-source', then it's fully indexed by Fabric, and can be searched semantically (by meaning, and the contents of the file).

***

## How does Fabric's search work?

* Just describe what you are looking for in your own words.
* The more descriptive you can be, the more likely that the search will be able to match the ideas, concepts and themes in your search.
* If you want an exact match, just wrap the query in quotes. For example: "research". This will only return results that specifically include that word.
* Fabric's search combines multiple factors to improve relevance beyond just the query itself, so that your searches find what you are looking for, despite the huge volume of data.

***

## How to use AI similarity search

* From any opened file inside Fabric, click the `⋮` (3 dots) button and click `Explore similar files`.
* Fabric will use that file as the search query and find the closest matches to it. A new way to explore your data via similarity.

***

## Search by color

* Use a color picker to find images, designs, and documents containing specific colors - perfect for brand consistency.
* Just click the color wheel in the search overlay, and then paste in a Hex colour, or drag within the color picker.
* Fabric will find the closest matching colored media for you.

***

## Reverse image search

* Upload an image and Fabric will find similar images and content. Reverse image search for your library of content.
* Just open the search view and click the <Icon icon="camera" /> camera icon to upload an image for the query.

***

## Media content search

* Just ask Fabric Assistant and it will search the transcript of audio, video files, and YouTube, finding specific timeframes within the media for you to hop straight into.
* Just click the timestamp to hop to that location in the media.
* No more hunting around for specific scene.

***

## Search filters

* A search doesn't need to include text, you can start just from file attributes, such as any tags, parent folders, or the kind of file that it is.
* You can also apply these filters to any other type of query, narrowing the results down.

***

## Search UI controls

You can customize Fabric's search view to your needs.

**There are multiple view modes in search:**

* Grid – shows a grid, ideal for most content
* Masonry – shows vertical columns, ideal for image/visual content
* Detailed – shows a list of deeper information with medium thumbnails
* List – granular information with customizable columns of key information

**Other:**

* You can zoom in or out as needed, making it easier to evaluate the content you are looking for.
* You can also toggle on and off visual attributes about grid or masonry cards, as well as whether the external search panel is visible.