Instagram Explore AI system

UPDATED JUL 30, 2025

The content you see on your Instagram Explore 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 Instagram Explore

When you view and interact with Instagram, one of the underlying AI systems fetches media (photo and video) based on user's engagement history and delivers media according to user's preferences.

How Instagram Explore works

The AI system behind Explore fetches quality and personalized media through three stages: retrieval, early-stage ranking, and late-stage ranking.

Retrieval
The retrieval stage is responsible for selecting a set of candidate items (in this case, photos and videos) that are relevant to the user's interests. The system uses a variety of techniques to retrieve candidates, including item collaborative filtering, personalized PageRank, and two tower sparse network sourcing. The fetched media contain sources from author-based sources (media from authors that you've engaged with) and media-based sources (media similar to media you've engaged with). At the end of the retrieval stage, up to 1500 media are fetched.
Early Stage Ranking
The early stage ranking stage is responsible for narrowing down the set of candidate items selected in the retrieval stage to a smaller set of the most promising candidates. This stage involves a two-tower neural network that uses both media features and user features to calculate the similarity between them. At the end of this stage, the top 100 media are passed to the next stage.
Late Stage Ranking
The late stage ranking stage is responsible for generating a final list of recommended items for the user. This stage typically involves applying multi-task multi-label neural network (a more complex machine learning model) to rank the remaining candidate items based on their likelihood of user engagement such as like and save. The final list of recommended items is then presented to the user in the Explore grid.

How to customize what you see

Your experience on Instagram Explore 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.

Not Interested

If you don’t want to see more of a certain type of content, you can select “Not Interested” in the three-dot overflow menu on an individual post. The system will attempt to filter out similar content in the future.

Click on "Like" to signal your interest in a post. The recommender system will recommend media similar to the ones you've liked

Report

If you see content you think goes against Instagram's Community Guidelines, you can report it.

See non-personalized content

To see posts and reels that aren't chosen or ordered in a way that's personalized specifically to you, click "Not personalized" in the Explore filter. 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.

Recommender One

Recommender Two

Recommender Three

Recommender Four

How likely you are to follow the author of a post