Skip to main content

Getting Started with Data Ingestion

Ingesting data is the first step in using Agentic RAG effectively after you have created your knowledgebox. This step is essential for enabling advanced search capabilities on your data.

Agentic RAG offers several methods for data ingestion:

  • Programmatically: Use the REST API or the Python SDK or the JavaScript SDK.
  • From a Terminal: Use the CLI.
  • Agentic RAG Dashboard: Upload files manually or sync a local folder or a sitemap with the Sync Agent through the web application.

Repeatable vs. One-Time Ingestions​

  • Repeatable Ingestions: If you need to ingest data regularly (e.g., daily, hourly), use the REST API or SDKs to automate the process. You can also use the Sync Agent to monitor third-party storage services for new files and sync them automatically.
  • One-Time Ingestions: For a single ingestion, use the Agentic RAG Dashboard web application or the CLI.
note

Storage limits

Depending on your account tier, the storage limit will differ. When the limit is reached, the /upload endpoint will return an error.

Data residency

When creating a Knowledge Box, you decide the geographical location of the data that will be stored in this Knowledge Box. Currently, Agentic RAG offers 2 regions: Europe and USA.