Zero data retention (ZDR)
Zero data retention (ZDR) means the AI provider keeps no copy of prompts or outputs after the response is returned. Content is processed in memory to answer the request and is not written to storage or logs.
Why it matters for companies
Standard API terms often include a retention window. Anthropic's Claude Code documentation, for example, describes a standard 30-day retention period, with zero data retention available to qualified Enterprise accounts through the account team.
For work under client confidentiality or with personal data, a ZDR commitment takes stored copies of prompts out of the risk picture, which shortens the security review.
Example
A law firm summarizes client documents with an LLM. With ZDR on, neither the gateway nor the model provider keeps a copy of a document once its summary is returned.
How Standard Compute handles it
Turn on zero data retention for your company, a team or a key, and every request is routed only to inference endpoints that commit to keeping no copy of your data. Teams and keys inherit the requirement and can never switch it off underneath you.
We process your prompts in memory to serve each request and never write them to storage or logs.
Related pages
- LLM gateway privacy
- LLM APIs that don't train on your data
- Security
- Data Processing Agreement
- Enterprise AI glossary