BedrockEmbedding

​

Parameters

Parameter	Type	Default	Description
aws_access_key_id	Optional[str]	None	AWS access key ID
aws_secret_access_key	Optional[str]	None	AWS secret access key
aws_session_token	Optional[str]	None	AWS session token
region_name	str	"us-east-1"	AWS region
profile_name	Optional[str]	None	AWS profile name
model_name	str	"amazon.titan-embed-text-v1"	Bedrock embedding model
model_id	Optional[str]	None	Full Bedrock model ID (overrides model_name)
inference_profile	Optional[str]	None	Bedrock inference profile
enable_guardrails	bool	True	Enable Bedrock guardrails
guardrail_id	Optional[str]	None	Custom guardrail ID
enable_model_caching	bool	True	Enable model response caching
prefer_provisioned_throughput	bool	False	Prefer provisioned throughput models
enable_cloudwatch_logging	bool	True	Enable CloudWatch logging
log_group_name	Optional[str]	None	CloudWatch log group name

​

Functions

​

__init__

Initialize the BedrockEmbedding provider. Parameters:

• config (Optional[BedrockEmbeddingConfig]): Configuration object
• **kwargs: Additional configuration options

​

_setup_aws_session

Setup AWS session with proper credentials.

​

_setup_bedrock_client

Setup Bedrock runtime client.

​

supported_modes

Get supported embedding modes. Returns:

• List[EmbeddingMode]: List of supported embedding modes

​

pricing_info

Get AWS Bedrock embedding pricing. Returns:

• Dict[str, float]: Pricing information

​

get_model_info

Get information about the current Bedrock model. Returns:

• Dict[str, Any]: Model information

​

_prepare_titan_request

Prepare request body for Titan models. Parameters:

• texts (List[str]): List of texts to embed

Returns:

• Dict[str, Any]: Request body

​

_prepare_cohere_request

Prepare request body for Cohere models. Parameters:

• texts (List[str]): List of texts to embed

Returns:

• Dict[str, Any]: Request body

​

_prepare_request_body

Prepare request body based on the model provider. Parameters:

• texts (List[str]): List of texts to embed
• mode (EmbeddingMode): Embedding mode

Returns:

• Dict[str, Any]: Request body

​

_extract_embeddings

Extract embeddings from response based on model type. Parameters:

• response_body (Dict[str, Any]): API response body
• num_texts (int): Number of texts

Returns:

• List[List[float]]: List of embedding vectors

​

_embed_batch

Embed a batch of texts using AWS Bedrock. Parameters:

• texts (List[str]): List of text strings to embed
• mode (EmbeddingMode): Embedding mode for optimization

Returns:

• List[List[float]]: List of embedding vectors

​

validate_connection

Validate Bedrock connection and model access. Returns:

• bool: True if connection is valid

​

get_aws_info

Get AWS-specific configuration information. Returns:

• Dict[str, Any]: AWS configuration information

​

get_cost_estimate

Get detailed cost estimation for current usage. Returns:

• Dict[str, Any]: Cost estimation

​

list_available_models

List available embedding models in Bedrock. Returns:

• List[Dict[str, Any]]: List of available models

​

close

Clean up AWS Bedrock connections and resources.

​

create_titan_embedding

Create AWS Titan embedding provider. Parameters:

• region_name (str): AWS region
• model_version (str): Model version
• **kwargs: Additional configuration options

Returns:

• BedrockEmbedding: Configured BedrockEmbedding instance

​

create_cohere_embedding

Create Cohere embedding provider. Parameters:

• language (str): Language for embedding
• region_name (str): AWS region
• **kwargs: Additional configuration options

Returns:

• BedrockEmbedding: Configured BedrockEmbedding instance