Create a Vector Space

Knowledge page listing your vector spaces with indexing and lookup metrics

A vector space is a knowledge base that stores document embeddings for semantic search. Agents use vector spaces to perform retrieval-augmented generation (RAG) over your documents.

Prerequisites

• You are signed in to the XpressAI Platform.
• You have an active project.

Steps

Create vector space form with name and embedding model selection

1. Navigate to the Knowledge section from the sidebar.
2. Click Create.
3. Enter a name for the vector space.
4. Select an embedding model from the dropdown. The default model works well for most use cases.
5. Click Create.

info

The vector space is ready for uploads immediately after creation. No additional configuration is required.

tip

Choose a descriptive name that reflects the content you plan to store, such as "Product Documentation" or "Support Tickets Q1." This makes it easier to manage multiple vector spaces.

Verify

• The new vector space appears in the Knowledge list.
• You can select it and see the upload interface ready for use.