ParsingConfiguration
Settings for parsing document contents. By default, the service converts the contents of each document into text before splitting it into chunks. To improve processing of PDF files with tables and images, you can configure the data source to convert the pages of text into images and use a model to describe the contents of each page.
To use a model to parse PDF documents, set the parsing strategy to BEDROCK_FOUNDATION_MODEL and
specify the model or inference profile to use by ARN. You can also override the default parsing prompt with instructions for how
to interpret images and tables in your documents. The following models are supported.
-
Anthropic Claude 3 Sonnet -
anthropic.claude-3-sonnet-20240229-v1:0 -
Anthropic Claude 3 Haiku -
anthropic.claude-3-haiku-20240307-v1:0
You can get the ARN of a model with the ListFoundationModels action. Standard model usage charges apply for the foundation model parsing strategy.
Contents
- parsingStrategy
-
The parsing strategy for the data source.
Type: String
Valid Values:
BEDROCK_FOUNDATION_MODELRequired: Yes
- bedrockFoundationModelConfiguration
-
Settings for a foundation model used to parse documents for a data source.
Type: BedrockFoundationModelConfiguration object
Required: No
See Also
For more information about using this API in one of the language-specific AWS SDKs, see the following:
