Configuring Vector Stores

Aizen provides vector store services. A vector store database is installed as part of the Aizen infrastructure components. See Installing the Infrastructure Components. Configure and start a vector store deployment to provide vector store services.

To configure a vector store on Aizen, follow these steps:

1. Log in to the Aizen Jupyter console. See Using the Aizen Jupyter Console.

2. Create an ML project if you have not already done so or set the current working project.

   create project <project name>

   or

   set project <project name>

3. Configure the vector store by running the configure vectorstore command:

   configure vectorstore

4. In the notebook, you will be guided through a template form with boxes and drop-down lists that you can complete to configure the vector store. You must add at least one embeddings model. For each embeddings model, select the embeddings model's Provider Name, enter the Embeddings Model name, and click the Add Model button. Set one of the embeddings models as the default model by checking the Set as default model check box.

5. Serve the vector store running the start vectorstore command. This command will schedule a job to deploy the vector store. Optionally, you can configure resources for the job by running the configure resource command. If you do not configure resources, default resource settings will be applied.

   configure resource
   start vectorstore <vectorstore name>

6. Check the status of the vector store's deployment job and obtain serving URLs:

   status vectorstore <vectorstore name>

7. The base URL in the status output supports a REST API that lists the vector stores that are currently being served. The endpoint URL in the status output supports vector store operations, such as creating store IDs, uploading documents, and retrieving semantically similar texts from a store ID. Both of these URLs provide the FastAPI docs, Redoc, and OpenAPI paths.

8. Use the endpoint URL to create a Store ID in the vector store and to upload documents to a Store ID.

9. Add a vector-store cloud provider with the base URL. You will need the provider name, vector store name, and the Store ID when configuring a RAG tool for an AI Agent.