Skip to main content
Ad Datasets is a REST API that exposes Meta Ad Library data in a shape that is easy to query from code and easy for LLM agents to call.

​
What You Get

  • Ads: creative, copy, metadata, and canonical fields optimized for querying and UI.
  • Brands: page-level metadata, counts, and recent ads.

​
Key Concepts

​
Canonical Fields

Canonical ad fields are cleaned and template-free where applicable (safe for UI/search). Full-fidelity raw GraphQL payloads are stored externally and are not inlined in the public API.

​
Summary-First Payloads

List and search endpoints return ad summaries by default to keep responses fast. Fetch a single ad by ad_archive_id for the full object.

​
Brands vs Creators

Every ad has a brand (who paid). Partnership/whitelisted ads also have a creator (whose identity appears in the creative). The creator_confidence field records the detection method.

​
Next Steps