Zero data retention

Together AI supports zero data retention (ZDR) at the organization level. With ZDR enabled, the prompts you send and the outputs models return are not stored by Together, and are not used for any secondary purpose.

​

Enable ZDR

ZDR is not enabled by default. To enable it:

1. Open the Privacy section of Organization Settings.
2. Set Store prompts and model responses to No.

Turning off prompt storage also turns off passthrough models automatically, because passthrough requires retained prompts. Non-passthrough models keep working as normal. A similar storage toggle exists on your personal account profile, but it only applies to traffic sent under your personal account outside an organization. When a request uses an organization’s API key, the organization setting is what applies.

​

Default behavior

Unless your organization enables ZDR, Together stores the prompts you send and the responses models return, and may use them for product improvements. This data will never be shared with third parties, and you can delete it at any time. Two related settings are independent of ZDR:

• Training: Use of your data for training models is a separate opt-in and is off by default. Together does not train on your data without your explicit consent, whether or not ZDR is enabled.
• Passthrough models: Models that forward requests to a third-party provider are allowed by default. Data sent to them is handled under that provider’s own policy.

​

What ZDR covers

With ZDR enabled, the content you submit to Together’s inference services, including text, images, and prompts, and any outputs returned to you are not stored, retained, or used for model training, product improvements, or any secondary purpose, except as needed to provide the service to you. Request content may pass through volatile memory and temporary caches while your request is served, but it is not persisted to non-volatile storage.

​

What Together retains with ZDR enabled

ZDR applies to the content of your requests. Together still retains the data it needs to run your account:

• Request metadata: Usage records such as token counts, model IDs, and timestamps, used for billing and the usage dashboard. Metadata does not include prompt or output content.
• Account data: Profile, organization, and billing information, handled per the privacy policy.
• Data you explicitly upload: Files you upload for fine-tuning or the batch API are stored so your jobs can run, and remain stored until you delete them through the API, CLI, or web console.

​

Enterprise deployments

For data-residency, regulatory, or compliance requirements, Together supports private networking and VPC-based deployments, including in EU regions, on dedicated endpoints. Contact us to discuss the right setup for your workload.

​

Privacy requests

For data access or deletion requests beyond what’s covered above, email [email protected]. The privacy policy is the authoritative legal statement of Together’s data practices.

​

Next steps