Namespaces
Isolated multi-tenant environments with dedicated vector storage, access controls, and complete data separation
Why do anything?
Multi-tenant applications need complete data isolation. Without namespaces, customer data can leak across boundaries, creating security and compliance risks.
Why now?
Enterprise AI applications demand SOC 2-grade controls and data residency. Manual tenant isolation is error-prone and doesn't scale.
Why this feature?
Each namespace maps to a dedicated Mixpeek Vector Store namespace with 12 protected payload indexes. Full cascade deletion ensures no orphaned data. Supports cross-namespace migration for tenant onboarding.
How It Works
Namespaces provide tenant isolation by mapping each namespace to a dedicated Mixpeek Vector Store namespace. All documents within a namespace share the same vector storage while maintaining complete separation from other namespaces.
Namespace Creation
Generate namespace_id, create the Mixpeek Vector Store namespace with the same name
Index Configuration
Set up 12 protected payload indexes for _internal fields
Access Control Setup
Configure namespace-scoped API keys and permissions
Cascade Deletion
12-step process: webhooks → alerts → taxonomies → clusters → retrievers → collections → batches → documents → buckets → secrets → namespace
Why This Approach
One Mixpeek Vector Store namespace per Mixpeek namespace ensures complete vector isolation. Protected indexes guarantee system field queryability. Cascade deletion prevents orphaned resources.
Where This Is Used
Recent updates
Full changelog- Sep 27, 2026Deploying a template copies all of it, and a copy in progress says soCloning a namespace could stop copying at an empty page of results on a shared store, so a Creative DNA workspace opened with 1 of its 1,583 documents. The clone now follows the cursor to the end. A clone whose worker stops now reads failed instead of copying indefinitely, and a retriever that is still being copied from its template answers with a clear still-copying status and where to watch progress, where it used to return a validation error.
- Sep 26, 2026The namespace API no longer returns Qdrant fieldsMVS is now the only vector store. qdrant_status, backend_compatibility and the qdrant_* infrastructure fields are gone from namespace responses, and the SDKs were regenerated to match. document_count and namespace_ready carry what qdrant_status reported. An older generated client that requires qdrant_collection will fail to parse a namespace that has an infrastructure block, so regenerate or upgrade it.
- Sep 25, 2026Namespace stats stop reporting 0 documents, and renaming a standalone namespace returns 200After an idle period, the first stats call for a namespace could report 0 documents while its counts were being refreshed. Stats now wait for the real count. Renaming a standalone namespace saved the new name but answered 500, and the namespace then could not be found by its id. The rename answers 200 and the id keeps working.
- Sep 25, 2026Namespace snapshots include triggers and alertsRestoring a namespace from a snapshot brought back its data and left out its automation: triggers, alerts and their execution history were not in the snapshot. Snapshots now carry all five, so a restore returns the namespace with its automation in place.