Data Processing & Retention

For content you submit, you are the data controller and Modelplex is a processor acting on your instructions (your API calls). We are, in turn, a controller for the account and billing metadata we collect to run the service.

Prompts and outputs are forwarded to the upstream provider you target to fulfill a request. Retention of that content is governed by the provider’s policy — some offer zero-data-retention or no-training options, others do not. We do not store prompt or completion content; we store only usage metadata (model, counts, timestamps). Generated media is re-hosted to our object storage and served via short-TTL signed URLs. A current list of sub-processors (model providers, payment, hosting, storage) is available on request.

Requests may be processed in the region of the selected provider, which can be outside your country. Where required, transfers rely on standard contractual clauses or equivalent safeguards.

• Prompt/output content: not stored by Modelplex (provider-dependent upstream).
• Usage metadata & ledger: retained for billing, audit, and legal obligations.
• Generated media: retained in object storage until you delete it.
• Account data: deleted on organization deletion (subject to legal holds).

For a DPA, sub-processor list, or a data deletion/export request, contact support@modelplex.ai.