Create Vector Stores
Vector Stores are AI-powered search indexes that organize your files for semantic search. Choose the right configuration for your use case, from temporary development environments to public knowledge bases.
Basic Creation
Create a new Vector Store with default settings:
When you create a Vector Store, you get a Vector Store object with a unique ID (UUID). You can use either the ID or the name you provide as Vector Store identifiers in all operations.
For complete details on the Vector Store object structure, see Data Models.
Configuration Options
Name (Vector Store Identifier)
The Vector Store name serves as both a human-readable identifier and works exactly like the Vector Store ID. Both are Vector Store identifiers that can be used interchangeably in all operations.
Name Properties:
- Must be unique within your organization
- Can be updated at any time
- Both name and Vector Store ID work as
vector_store_identifiers
Expiration Policies
Set automatic cleanup policies to manage Vector Store lifecycle based on activity:
Expiration Behavior:
last_active_at
- Activity-based expiration
- Timer resets on:
- add files
- delete files
- search operations
- Best for: Development environments, temporary projects, demos
Default Behavior:
- Vector stores never expire if
expires_after
is not set - No automatic cleanup unless explicitly configured
Example Use Cases:
- Development: Set 7-day expiration for testing environments
- Demos: Set 30-day expiration for temporary showcases
- Production: No expiration for permanent knowledge bases
Public Access
Control read access to your Vector Store:
Access Levels:
Public (is_public=True
)
- Other Mixedbread users can search your Vector Store
- Great for sharing documentation, FAQs, or knowledge bases
- Billing: Search costs are paid by the person searching (they need an API key)
- Use cases: Product docs, help centers, research papers
Private (is_public=False
)
- Only your organization can access the Vector Store
- Billing: All costs are paid by your organization
- Use cases: Internal documents, customer data, proprietary content
Next Steps
Now that you've created your Vector Store, learn how to manage it:
- Manage Vector Stores: Update, retrieve, list, and delete Vector Stores
- Ingest: Add files to your Vector Store
- Search: Query your Vector Store content
- Data Models: Understand Vector Store structure
Last updated: July 15, 2025