Bedrock to Sharepoint Knowledge Base Costs

When using Amazon Bedrock Knowledge Bases with SharePoint as a data source, there is no direct charge for the Knowledge Bases feature itself. However, you will incur costs for two main components:

1. The embedding model used to process your data: You'll be charged based on the specific embedding model you select (such as Embed English v3) and the volume of tokens processed.

2. The vector database that stores your embeddings: If you're using the recommended "Quick create a new vector store" option, this typically provisions an Amazon OpenSearch Serverless instance. The costs for this will depend on the storage and compute resources consumed.

There isn't a simple "per GB" pricing model for ingesting SharePoint data into a Bedrock Knowledge Base. The total cost will depend on:

• The size and complexity of your SharePoint documents
• How frequently you sync or update the data
• The specific embedding model you choose
• The vector database configuration

To avoid unnecessary costs, remember to delete any knowledge bases you create for experimentation purposes when you're done with them.

For a more precise cost estimate, you would need to consider the specific models you plan to use and the expected query volume against your knowledge base.
Sources
Pricing for AWS Bedrock knowledge bases | AWS re:Post
Integrate Amazon Bedrock Knowledge Bases with Microsoft SharePoint as a data source | AWS Machine Learning Blog
Cost - Generative AI Application Builder on AWS