AWS::Bedrock::Prompt PromptModelInferenceConfiguration - Amazon CloudFormation
SyntaxProperties
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).

AWS::Bedrock::Prompt PromptModelInferenceConfiguration

Contains inference configurations related to model inference for a prompt. For more information, see Inference parameters.

Syntax

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

JSON

{
  "MaxTokens" : Number,
  "StopSequences" : [ String, ... ],
  "Temperature" : Number,
  "TopK" : Number,
  "TopP" : Number
}

YAML

  MaxTokens: Number
  StopSequences: 
    - String
  Temperature: Number
  TopK: Number
  TopP: Number

Properties

MaxTokens

The maximum number of tokens to return in the response.

Required: No

Type: Number

Minimum: 0

Maximum: 4096

Update requires: No interruption

StopSequences

A list of strings that define sequences after which the model will stop generating.

Required: No

Type: Array of String

Minimum: 0

Maximum: 4

Update requires: No interruption

Temperature

Controls the randomness of the response. Choose a lower value for more predictable outputs and a higher value for more surprising outputs.

Required: No

Type: Number

Minimum: 0

Maximum: 1

Update requires: No interruption

TopK

The number of most-likely candidates that the model considers for the next token during generation.

Required: No

Type: Number

Minimum: 0

Maximum: 500

Update requires: No interruption

TopP

The percentage of most-likely candidates that the model considers for the next token.

Required: No

Type: Number

Minimum: 0

Maximum: 1

Update requires: No interruption