View a markdown version of this page

AWS::BedrockAgentCore::Harness HarnessLiteLlmModelConfig - Amazon CloudFormation
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

This is the new Amazon CloudFormation Template Reference Guide. Please update your bookmarks and links. For help getting started with CloudFormation, see the Amazon CloudFormation User Guide.

AWS::BedrockAgentCore::Harness HarnessLiteLlmModelConfig

Configuration for a LiteLLM model provider, enabling connection to third-party model providers.

Syntax

To declare this entity in your Amazon CloudFormation template, use the following syntax:

JSON

{ "AdditionalParams" : {Key: Value, ...}, "ApiBase" : String, "ApiKeyArn" : String, "MaxTokens" : Integer, "ModelId" : String, "Temperature" : Number, "TopP" : Number }

YAML

AdditionalParams: Key: Value ApiBase: String ApiKeyArn: String MaxTokens: Integer ModelId: String Temperature: Number TopP: Number

Properties

AdditionalParams

Provider-specific parameters passed through to the model provider unchanged.

Required: No

Type: Object

Update requires: No interruption

ApiBase

The base URL for the model provider's API endpoint.

Required: No

Type: String

Minimum: 1

Maximum: 16383

Update requires: No interruption

ApiKeyArn

The ARN of the API key in AgentCore Identity for authenticating with the model provider.

Required: No

Type: String

Pattern: ^arn:aws:bedrock-agentcore:[a-z0-9-]+:[0-9]{12}:token-vault/[a-zA-Z0-9-.]+/apikeycredentialprovider/[a-zA-Z0-9-.]+$

Update requires: No interruption

MaxTokens

The maximum number of tokens to allow in the generated response per iteration.

Required: No

Type: Integer

Minimum: 1

Update requires: No interruption

ModelId

The LiteLLM model identifier (e.g., "anthropic/claude-3-sonnet").

Required: Yes

Type: String

Update requires: No interruption

Temperature

The temperature to set when calling the model.

Required: No

Type: Number

Minimum: 0

Maximum: 2

Update requires: No interruption

TopP

The topP set when calling the model.

Required: No

Type: Number

Minimum: 0

Maximum: 1

Update requires: No interruption