Package org.springframework.ai.bedrock.titan

Class BedrockTitanEmbeddingClient

java.lang.Object

org.springframework.ai.embedding.AbstractEmbeddingClient

org.springframework.ai.bedrock.titan.BedrockTitanEmbeddingClient

All Implemented Interfaces:

EmbeddingClient, ModelClient<EmbeddingRequest,EmbeddingResponse>

public class BedrockTitanEmbeddingClient
extends AbstractEmbeddingClient

EmbeddingClient implementation that uses the Bedrock Titan Embedding API. Titan Embedding supports text and image (encoded in base64) inputs. Note: Titan Embedding does not support batch embedding.

Since:

0.8.0

Author:

Christian Tzolov

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static enum	BedrockTitanEmbeddingClient.InputType

Field Summary

Fields inherited from class org.springframework.ai.embedding.AbstractEmbeddingClient

embeddingDimensions

Constructor Summary

Constructors

Constructor	Description
BedrockTitanEmbeddingClient(TitanEmbeddingBedrockApi titanEmbeddingBedrockApi)

Method Summary

Modifier and Type	Method	Description
EmbeddingResponse	call(EmbeddingRequest request)	Executes a method call to the AI model.
int	dimensions()
List<Double>	embed(Document document)	Embeds the given document's content into a vector.
BedrockTitanEmbeddingClient	withInputType(BedrockTitanEmbeddingClient.InputType inputType)	Titan Embedding API input types.

Methods inherited from class org.springframework.ai.embedding.AbstractEmbeddingClient

dimensions

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.springframework.ai.embedding.EmbeddingClient

embed, embed, embedForResponse

Constructor Details

BedrockTitanEmbeddingClient

public BedrockTitanEmbeddingClient(TitanEmbeddingBedrockApi titanEmbeddingBedrockApi)

Method Details

withInputType

public BedrockTitanEmbeddingClient withInputType(BedrockTitanEmbeddingClient.InputType inputType)

Titan Embedding API input types. Could be either text or image (encoded in base64).

Parameters:

inputType - the input type to use.

embed

public List<Double> embed(Document document)

Description copied from interface: EmbeddingClient

Embeds the given document's content into a vector.

Parameters:

document - the document to embed.

Returns:

the embedded vector.

call

public EmbeddingResponse call(EmbeddingRequest request)

Description copied from interface: ModelClient

Executes a method call to the AI model.

Parameters:

request - the request object to be sent to the AI model

Returns:

the response from the AI model

dimensions

public int dimensions()

Specified by:

dimensions in interface EmbeddingClient

Overrides:

dimensions in class AbstractEmbeddingClient

Returns:

the number of dimensions of the embedded vectors. It is generative specific.