Backfrom Meta’s own policies · verified 2026-09-02 · some facts not published

Meta: data residency & retention

Headquarters
🇺🇸 United States (US)
Where inference runs
Not disclosed
Meta does not publish where Llama API inference runs; it states Llama API data is stored separately from other Meta product data, but names no regions.
Region pinning
Not disclosed
No region-pinning controls disclosed for the Llama API developer platform.
Prompt retention
Not disclosed
No specific prompt-retention window published for the Llama API; developer terms page could not be rendered (JS-only) and the LlamaCon announcement gives commitments but no retention schedule.
Trains on API data
No
Meta's own announcement: 'We do not use your prompts or model responses to train our AI models.' Also states data is not used for ads and models you fine-tune are portable.
Zero data retention
Not disclosed
No formal zero-data-retention tier disclosed.

Compiled from Meta’s published privacy policy, terms, and documentation on 2026-09-02. “Not disclosed” means Meta publishes nothing on the point. Policies change without notice, so verify against the sources above before relying on this for compliance decisions.

Standard Compute

Plug in Standard Compute wherever you need intelligence.

One key. Every frontier model. A bill that stays put.

7-day fair refund Cancel anytime Billing by Stripe