Facebook Search AI system

UPDATED SEP 25, 2025
The content you see on your Facebook Search is selected, ranked and delivered to you by an artificial intelligence (AI) system. Within one AI system, multiple machine learning models work together to deliver your experience. These models and their input signals are dynamic and they change frequently as the system learns and improves over time.

Overview of Facebook Search

When you view and interact with Facebook, one of the underlying AI systems delivers results when you search for content. These are the results you see when you use the main search bar on Facebook Feed.
How Facebook Search works
The AI system behind Search allows users to quickly find people, posts, Pages, Groups, and other content across the Facebook app. It leverages advanced algorithms to deliver relevant results, helping users discover and connect with information and communities that matter to them. Here's how it works:
  1. Gather inventory
    First, the system collects all eligible search results to rank for you. This content includes posts, friends, groups, events, Pages and other results that are relevant to the words you searched for.
  2. Score results
    Next, the system scores each search result for a variety of factors, such as the type of content and how well it matches what you tend to interact with. The system narrows the pool of eligible content to the most relevant search results for you.
  3. Apply additional filters
    Then, integrity processes are applied to all results. These determine which integrity measures, if any, need to be applied to the results selected for ranking.
  4. Rank results by score
    Finally, the system puts all the results in order by score. Results that the system predicts will be most valuable or relevant for you are shown to you first.
How to customize what you see
Your experience on Facebook Search is personalized based on your activity, and you have options to control or customize what you see. Below, we describe how to do this with different in-product features. Options shown here may not be available to everyone.
Tabs and filtering
You can narrow search results in two ways:
After you perform a search, you can select a category from the tabs that appear, such as posts, people or groups. You’ll then see search results from only that category.
You can also filter search results by first clicking on a tab and then selecting options to filter results based on various common attributes, such as location or date posted.
Clear recent search history
You can clear all your recent searches or remove a search from your history. Searches you clear may reappear in your history after you search for them again.
See non-personalized search results
To see search results that are not personalized specifically to you, use the search feature in the Feeds tab. By doing so, your search results will be based on the search term you enter. Learn more.
How the AI delivers content to you
We want you to see content you enjoy and find interesting. To achieve this, the AI system has models that help it make predictions about content you'll find most relevant and valuable. These prediction models use underlying input signals to help select content you're most likely to engage with.
Below are some of the significant predictions–and input signals that inform them–that we use in this AI system.
How likely you are to interact with a reel in a search result, such as by clicking to view it
Signals influencing this prediction include:
  • Post number of likes
  • Predicted post language
  • ML model based prediction if query is looking for news
  • Time between when post is create to now
  • Content side click through rate
How likely you are to interact with a photo in a search result, such as by clicking to view or liking it
Signals influencing this prediction include:
  • Predicted post language
  • Time between when post is create to now
  • ML model based prediction if query is looking for news
  • Post number of likes
  • Content side click through rate
How likely you are to interact with a video in a search result, such as by clicking to view, liking or commenting on it
Signals influencing this prediction include:
  • Post number of likes
  • Time between when post is create to now
  • ML model based prediction if query is looking for news
  • Predicted post language
  • Content side click through rate
How likely you are to interact with a post in a search result, such as by commenting, clicking to view or liking it
Signals influencing this prediction include:
  • Predicted post language
  • ML model based prediction if query is looking for news
  • Time between when post is create to now
  • Post number of likes
  • Content side click through rate