Data retention - verified
Does Cerebras store your API prompts?
Default retention: not retained by default. Trains on your API data: Not stated. Zero data retention: on by default (no separate program; the privacy policy says inputs and outputs are not retained). Regional processing: none documented.
Read from cerebras.ai, inference-docs.cerebras.ai. Compare every AI API's data retention
- Trains on data
- Not stated
- ZDR
- On by default
- Regions
- None
- Verified
- 2026-10-18
How long are prompts kept by default?
- Default retention
Not retained by default
Inputs and outputs of inference services not retained; associated logs deleted when no longer necessary.
“We do not retain inputs and outputs associated with our training, inference and chatbot Services. We delete logs associated with our training, inference and chatbot Services when they are no longer necessary to provide services to you”
cerebras.ai - Duration
- Not stated
- Purpose
- Not stated
Does it train on your API data?
Not stated
No explicit training statement found in the privacy policy or inference docs fetched.
No statement found on the pages checked.
Zero data retention (ZDR)
- Availability
On by default: prompts are not retained unless you opt in
No separate program; the privacy policy says inputs and outputs are not retained
Recorded as: No separate ZDR program documented; privacy policy states inference inputs/outputs are not retained
- Who can get it
Default behaviour per privacy policy (no retention of inputs/outputs); no separate ZDR program page found.
- Not covered
Batch files and results retained 7 days.
- From the docs
“Yes, it is fully ZDR-compliant. All cached context remains ephemeral in memory and never persisted.”
inference-docs.cerebras.ai
Features kept longer
Stateful features hold your content until a job finishes, a window expires or you delete it, often regardless of the default above.
| Feature | Retention | The docs say |
|---|---|---|
| Batch results | 7 days after completion | “Results retained for 7 days after completion”inference-docs.cerebras.ai |
| Uploaded files (purpose=batch) | 7 days by default | “Files with purpose=batch expire after 7 days by default.”inference-docs.cerebras.ai |
| Prompt cache (in memory) | Guaranteed TTL 5 minutes, up to 1 hour | “Guaranteed TTL is 5 minutes, but up to 1 hour max depending on system load.”inference-docs.cerebras.ai |
Where is your data stored and processed?
- Regional processing
None documented
- Stated location
United States and other countries
Recorded as: United States and other countries outside the United States
- How to choose
- Not stated
- What it covers
Both (stored and processed).
- Summary
No region selection documented for the API. Privacy policy: data may be stored and processed in the US and other countries unless a separate agreement says otherwise. Cache is colocated with the serving data center.
- From the docs
“Except as otherwise agreed to in a separate agreement with you, your data may be stored and processed in the United States and other countries outside the United States, which may have data protection laws that differ from the laws in your country.”
cerebras.ai “Cached tokens are stored in key-value stores colocated in the same data center as the model instance serving your traffic.”
inference-docs.cerebras.ai
Regional endpoints
Regional hostnames: Not stated
Certifications
Certifications on the pages checked: Not stated
Research notes
- Retention language comes from the general Cerebras privacy policy (modified 2025-03-19 per page metadata), not an API-specific data page. Cerebras marketing blog posts mention US data centers and zero data retention, but those are not docs and were not used.
FAQ
Does Cerebras store your API prompts?
Not retained by default. Inputs and outputs of inference services not retained; associated logs deleted when no longer necessary. The provider says: "We do not retain inputs and outputs associated with our training, inference and chatbot Services. We delete logs associated with our training, inference and chatbot Services when they are no longer necessary to provide services to you"
Does Cerebras train on your API data?
Not stated. No explicit training statement found in the privacy policy or inference docs fetched.
Does Cerebras offer zero data retention?
On by default (no separate program; the privacy policy says inputs and outputs are not retained). Default behaviour per privacy policy (no retention of inputs/outputs); no separate ZDR program page found.
Where does Cerebras process your data?
Regional processing options: None documented. Stated processing location: United States and other countries.
Sources
Last verified from Cerebras's own pages below. Quotes are verbatim.
This page summarizes what each provider's own pages say. It is not legal advice; your contract, data processing addendum and the provider's current terms govern what happens to your data.
Newsletter
Know when a provider changes its data terms
One email per week. New model releases, pricing and policy changes, and what they mean for what you build.