OpenAIAssistantInvocationOptions Class

Definition

Namespace:: Microsoft.SemanticKernel.Agents.OpenAI

Assembly:: Microsoft.SemanticKernel.Agents.OpenAI.dll

Package:: Microsoft.SemanticKernel.Agents.OpenAI v1.45.0-preview

Important

Some information relates to prerelease product that may be substantially modified before it’s released. Microsoft makes no warranties, express or implied, with respect to the information provided here.

Caution

Use RunCreationOptions to specify assistant invocation behavior.

Defines per-invocation execution options that override the assistant definition.

[System.Obsolete("Use RunCreationOptions to specify assistant invocation behavior.")]
public sealed class OpenAIAssistantInvocationOptions

[<System.Obsolete("Use RunCreationOptions to specify assistant invocation behavior.")>]
type OpenAIAssistantInvocationOptions = class

Public NotInheritable Class OpenAIAssistantInvocationOptions

Inheritance: Object
OpenAIAssistantInvocationOptions

Attributes: ObsoleteAttribute

Remarks

This class is not applicable to AgentChat usage.

Constructors

OpenAIAssistantInvocationOptions()

Properties

AdditionalInstructions	Gets the additional instructions.
AdditionalMessages	Gets additional messages to add to the thread.
EnableCodeInterpreter	Gets a value that indicates if the code_interpreter tool is enabled.
EnableFileSearch	Gets a value that indicates if the file_search tool is enabled.
EnableJsonResponse	Gets a value that indicates if the JSON response format is enabled.
MaxCompletionTokens	Gets the maximum number of completion tokens that can be used over the course of the run.
MaxPromptTokens	Gets the maximum number of prompt tokens that can be used over the course of the run.
Metadata	Gets a set of up to 16 key/value pairs that can be attached to an agent, used for storing additional information about that object in a structured format.
ModelName	Gets the AI model targeted by the agent.
ParallelToolCallsEnabled	Gets a value that indicates whether parallel function calling is enabled during tool use.
Temperature	Gets the sampling temperature to use, between 0 and 2.
TopP	Gets the probability mass of tokens whose results are considered in nucleus sampling.
TruncationMessageCount	Gets the number of recent messages that the thread will be truncated to.

Applies to