Chart · default retentionUpdated November 18, 2026
How long AI APIs keep your prompts, on a retention timeline
AI APIs keep your prompts anywhere from not at all to 55 days by default.
Silk Compute does not store your prompts or model outputs.
Default retention of API prompts and outputs
Not storedKept by defaultKept longer if flagged
Silk ComputeOpen-weight models
Zero data retention
abliteration.aiOpen-weight models
Never stored by default
FeatherlessOpen-weight models
Not logged
Unfil.AIOpen-weight models
Up to 24 hours
xAI APIGrok models
30 days
OpenAI APIGPT models
Up to 30 days
Claude APIClaude models
Within 30 daysUp to 2 years if flagged2 years if flagged
MorphOpen-weight models
Up to 30 days
Gemini APIGemini models
55 days
Mistral APIMistral models
No period published
PRIVATE AI APICC-BYRead November 18, 2026
Every bar and its source
| API | Default | The clause | Source |
|---|---|---|---|
| Silk ComputeOpen-weight models | Zero data retention | "Silk Compute does not store your prompts or model outputs." | silkcompute.com |
| abliteration.aiOpen-weight models | Never stored by defaultWeb search and fetch excluded | "Prompts, completions, and images are processed in memory and never stored by default. They are never used to train or fine-tune any model." | abliteration.ai |
| FeatherlessOpen-weight models | Not logged | "Featherless does not log chats, prompts, or completions sent through our API. Requests made via the API are processed in real time and are not stored on our servers." | featherless.ai |
| Unfil.AIOpen-weight models | Up to 24 hoursTemporary processing data | "We do not store your API prompts or responses after processing" | unfil.ai |
| xAI APIGrok models | 30 daysAbuse and misuse audits | "By default, all API requests and responses are stored on our servers (encrypted at rest) for 30 days for auditing purposes in the event of suspected abuse or misuse." | docs.x.ai |
| OpenAI APIGPT models | Up to 30 daysAbuse-monitoring logs | "By default, abuse monitoring logs are generated for all API feature usage and retained for up to 30 days, unless longer retention is required by law…" | developers.openai.com |
| Claude APIClaude models | Within 30 daysUp to 2 years if flagged | "For Anthropic API users, we automatically delete inputs and outputs on our backend within 30 days of receipt or generation…" | privacy.claude.com |
| MorphOpen-weight models | Up to 30 daysFree and paid tiers | "Free and paid tiers, minimal retention: We keep Inputs and Outputs only to operate the Services and your own request history, delete them within 30 days, and do not review them in the ordinary course." | morphllm.com |
| Gemini APIGemini models | 55 daysPrompts, context and output | "Google retains the following data for fifty-five (55) days…" | aistudio.google.com |
| Mistral APIMistral models | No period published | "Zero Data Retention (ZDR) is available only with pay-as-you-go and only for stateless API calls." | help.mistral.ai |
What the bars don't show
Four things can keep a prompt longer than its bar.
- Anthropic keeps flagged chats up to 2 years, says its privacy center, and OpenAI keeps images flagged as potential CSAM even under Zero Data Retention, says its data controls page.
- OpenAI's Responses API and xAI keep responses 30 days unless you send
store: false, and files, batches and threads stay until you delete them. - Google keeps prompts sent with Grounding with Google Search 30 days, and abliteration.ai's zero retention skips web search and fetch.
- Anthropic's Covered Models, such as Claude Fable 5.2, keep prompts 30 days on every platform.
How to read it
A ring at the left edge means the provider says it stores no prompts, and a dashed track means it publishes no period. To get a bar to zero, see how to get zero data retention, and for the full order, see the most private AI APIs.