CreateRunOptions interface
The details used when creating a new run of an agent thread.
Properties
| additional |
Additional instructions to append at the end of the instructions for the run. This is useful for modifying the behavior on a per-run basis without overriding other instructions. |
| additional |
Adds additional messages to the thread before creating the run. |
| assistant |
The ID of the agent that should run the thread. |
| instructions | The overridden system instructions that the agent should use to run the thread. |
| max |
The maximum number of completion tokens that may be used over the course of the run. The run will make a best effort
to use only the number of completion tokens specified, across multiple turns of the run. If the run exceeds the number of
completion tokens specified, the run will end with status |
| max |
The maximum number of prompt tokens that may be used over the course of the run. The run will make a best effort to use only
the number of prompt tokens specified, across multiple turns of the run. If the run exceeds the number of prompt tokens specified,
the run will end with status |
| metadata | A set of up to 16 key/value pairs that can be attached to an object, used for storing additional information about that object in a structured format. Keys may be up to 64 characters in length and values may be up to 512 characters in length. |
| model | The overridden model name that the agent should use to run the thread. |
| parallel |
If |
| response |
Specifies the format that the model must output. |
| stream | If |
| temperature | What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. |
| tool |
Controls whether or not and which tool is called by the model. |
| tools | The overridden list of enabled tools that the agent should use to run the thread. |
| topP | An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both. |
| truncation |
The strategy to use for dropping messages as the context windows moves forward. |
Property Details
additionalInstructions
Additional instructions to append at the end of the instructions for the run. This is useful for modifying the behavior on a per-run basis without overriding other instructions.
additionalInstructions?: null | stringProperty Value
null | string
additionalMessages
Adds additional messages to the thread before creating the run.
additionalMessages?: null | ThreadMessageOptions[]Property Value
null | ThreadMessageOptions[]
assistantId
The ID of the agent that should run the thread.
assistantId: stringProperty Value
string
instructions
The overridden system instructions that the agent should use to run the thread.
instructions?: null | stringProperty Value
null | string
maxCompletionTokens
The maximum number of completion tokens that may be used over the course of the run. The run will make a best effort
to use only the number of completion tokens specified, across multiple turns of the run. If the run exceeds the number of
completion tokens specified, the run will end with status incomplete. See incomplete_details for more info.
maxCompletionTokens?: null | numberProperty Value
null | number
maxPromptTokens
The maximum number of prompt tokens that may be used over the course of the run. The run will make a best effort to use only
the number of prompt tokens specified, across multiple turns of the run. If the run exceeds the number of prompt tokens specified,
the run will end with status incomplete. See incomplete_details for more info.
maxPromptTokens?: null | numberProperty Value
null | number
metadata
A set of up to 16 key/value pairs that can be attached to an object, used for storing additional information about that object in a structured format. Keys may be up to 64 characters in length and values may be up to 512 characters in length.
metadata?: null | Record<string, string>Property Value
null | Record<string, string>
model
The overridden model name that the agent should use to run the thread.
model?: null | stringProperty Value
null | string
parallelToolCalls
If true functions will run in parallel during tool use.
parallelToolCalls?: booleanProperty Value
boolean
responseFormat
Specifies the format that the model must output.
responseFormat?: null | AgentsApiResponseFormatOptionProperty Value
stream
If true, returns a stream of events that happen during the Run as server-sent events,
terminating when the Run enters a terminal state with a data: [DONE] message.
stream?: booleanProperty Value
boolean
temperature
What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
temperature?: null | numberProperty Value
null | number
toolChoice
Controls whether or not and which tool is called by the model.
toolChoice?: null | AgentsApiToolChoiceOptionProperty Value
null | AgentsApiToolChoiceOption
tools
The overridden list of enabled tools that the agent should use to run the thread.
tools?: null | ToolDefinition[]Property Value
null | ToolDefinition[]
topP
An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
We generally recommend altering this or temperature but not both.
topP?: null | numberProperty Value
null | number
truncationStrategy
The strategy to use for dropping messages as the context windows moves forward.
truncationStrategy?: null | TruncationObjectProperty Value
null | TruncationObject
