F3 Lightning: How Meta Accelerates Ads ML Features to Production at Scale

Meta detailed F3 Lightning, an extension of its Facebook Feature Framework (F3) that compiles the machine learning features behind ads ranking models into a single merged graph, a design Meta says cut execution computational costs. F3 serves requests that can require more than 20,000 features per request; Meta first presented F3 at AI @ Scale 2020 and published the F3 Lightning update on its Engineering at Meta blog on November 3, 2026. The post matters to teams running large-scale feature pipelines because it describes how Meta moves ads ML features into production.

Facebook Feature Framework F3 is Meta’s framework for authoring, compiling, and serving the machine learning features underlying ads ranking models, where a single request can require more than 20,000 features. We first presented F3 at AI @ Scale 2020. F3 reduced execution computational costs by compiling all features into a single merged graph, but that ... The post F3 Lightning: How Meta Accelerates Ads ML Features to Production at Scale https://engineering.fb.com/2026/11/03/ml-applications/f3-lightning-how-meta-accelerates-ads-ml-features-to-production/ appeared first on Engineering at Meta https://engineering.fb.com .