Perplexity API
Perplexity AI · Model API · US · docs.perplexity.ai
Last verified 2026-09-16
Entry
services/perplexity-api.yamlTraining on your data (default tier)
Zero data retention for the Chat Completions API; customer data not used for training
Retention
Only billing metadata kept (token counts, model, timestamp, API key); no prompt or response content
Data residency
Not stated; US company
Vendor data-handling terms
Training on your data (default tier)
Zero data retention for the Chat Completions API; customer data not used for training
“We do not retain data sent through the Chat Completions API, and we do not use customer data to train our models or for any purpose beyond processing the immediate request.”
Retention
Only billing metadata kept (token counts, model, timestamp, API key); no prompt or response content
“This billing metadata does not include any content from your prompts, responses, or other user data.”
Data residency
Not stated; US company
The API documentation names no processing region.
Domains and endpoints
observed means seen in Unseen deployments; vendor-documented means listed by the vendor. Vendors do not publish complete lists.
| Host | Role | Source |
|---|---|---|
| api.perplexity.ai | api | observed |
Assessment
API: low
Reasoning: The Sonar API behind Perplexity. Zero data retention and no training are stated in the developer documentation; region is not.
Flags: Zero data retention · Region not stated · Assessed 2026-09-16. The assessment is Unseen's; the terms above are the vendor's.
Changelog
- 2026-09-16Documented. Retention and training terms verified against the Perplexity API privacy documentation.
- 2026-09-15Listed from the Unseen catalogue with observed domains.
Corrections
Pull request on GitHub, or the form below. Changes are reviewed and recorded in the changelog.