Class ChatBedrockConverse

Integration with AWS Bedrock Converse API.

Example

import { ChatBedrockConverse } from "@langchain/aws";

const model = new ChatBedrockConverse({
  region: process.env.BEDROCK_AWS_REGION ?? "us-east-1",
  credentials: {
    secretAccessKey: process.env.BEDROCK_AWS_SECRET_ACCESS_KEY!,
    accessKeyId: process.env.BEDROCK_AWS_ACCESS_KEY_ID!,
  },
});

const res = await model.invoke([new HumanMessage("Print hello world")]);

Hierarchy (view full)

Implements

Constructors

constructor

Properties

client model region streamUsage streaming additionalModelRequestFields? endpointHost? guardrailConfig? maxTokens? temperature? topP?

Methods

bindTools getLsParams invocationParams

Constructors

constructor

Properties

client

client: BedrockRuntimeClient

model

model: string = "anthropic.claude-3-haiku-20240307-v1:0"

Model to use. For example, "anthropic.claude-3-haiku-20240307-v1:0", this is equivalent to the modelId property in the list-foundation-models api. See the below link for a full list of models.

Link

https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html#model-ids-arns

Default

anthropic.claude-3-haiku-20240307-v1:0

region

region: string

The AWS region e.g. us-west-2. Fallback to AWS_DEFAULT_REGION env variable or region specified in ~/.aws/config in case it is not provided here.

streamUsage

streamUsage: boolean = true

Whether or not to include usage data, like token counts in the streamed response chunks. Passing as a call option will take precedence over the class-level setting.

Default

true

streaming

streaming: boolean = false

Whether or not to stream responses

Optional additionalModelRequestFields

additionalModelRequestFields?: null | string | number | boolean | DocumentType[] | {
    [prop: string]: DocumentType;
}

Additional inference parameters that the model supports, beyond the base set of inference parameters that the Converse API supports in the inferenceConfig field. For more information, see the model parameters link below.

Type declaration

  • [prop: string]: DocumentType

Link

https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters.html

Optional endpointHost

endpointHost?: string

Override the default endpoint hostname.

Optional guardrailConfig

guardrailConfig?: GuardrailConfiguration

Configuration information for a guardrail that you want to use in the request.

Optional maxTokens

maxTokens?: number = undefined

Max tokens.

Optional temperature

temperature?: number = undefined

Temperature.

Optional topP

topP?: number

The percentage of most-likely candidates that the model considers for the next token. For example, if you choose a value of 0.8 for topP, the model selects from the top 80% of the probability distribution of tokens that could be next in the sequence. The default value is the default value for the model that you are using. For more information, see the inference parameters for foundation models link below.

Link

https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters.html

Methods

bindTools

  • bindTools(tools, kwargs?): Runnable<BaseLanguageModelInput, AIMessageChunk, this["ParsedCallOptions"]>

  • Parameters

    • tools: any[]
    • Optional kwargs: Partial<unknown>

    Returns Runnable<BaseLanguageModelInput, AIMessageChunk, this["ParsedCallOptions"]>

getLsParams

invocationParams

Settings

Member Visibility

Theme

On This Page

Generated using TypeDoc