class

OpenAI::Response

Inherits JSON::Serializable < Reference < Object

Constructors

new(pull : JSON::PullParser)

new(id : String = "", object : ResponseObject = ResponseObject::Response, status : ResponseStatus | Nil = nil, created_at : Float64 = 0.0, error : ResponseError = nil, incomplete_details : ResponseIncompleteDetails = nil, output : Array(OutputItem) = [] of OutputItem, instructions : ResponseInstructions = nil, output_text : ResponseOutputText | Nil = nil, usage : ResponseUsage | Nil = nil, parallel_tool_calls : Bool = false, conversation : ResponseConversation | Nil = nil)

new(*, __pull_for_json_serializable pull : JSON::PullParser)

Instance methods

conversation

conversation=(conversation : ResponseConversation | Nil)

created_at

Unix timestamp (in seconds) of when this Response was created.

created_at=(created_at : Float64)

Unix timestamp (in seconds) of when this Response was created.

error

error=(error : ResponseError)

id

Unique identifier for this Response.

id=(id : String)

Unique identifier for this Response.

incomplete_details

incomplete_details=(incomplete_details : ResponseIncompleteDetails)

instructions

instructions=(instructions : ResponseInstructions)

object

The object type of this resource - always set to response.

object=(object : ResponseObject)

The object type of this resource - always set to response.

output

An array of content items generated by the model.

The length and order of items in the output array is dependent on the model's response.
Rather than accessing the first item in the output array and assuming it's an assistant message with the content generated by the model, you might consider using the output_text property where supported in SDKs.

output=(output : Array(OutputItem))

An array of content items generated by the model.

The length and order of items in the output array is dependent on the model's response.
Rather than accessing the first item in the output array and assuming it's an assistant message with the content generated by the model, you might consider using the output_text property where supported in SDKs.

output_text

output_text=(output_text : ResponseOutputText | Nil)

parallel_tool_calls

Whether to allow the model to run tool calls in parallel.

parallel_tool_calls=(parallel_tool_calls : Bool)

Whether to allow the model to run tool calls in parallel.

status

The status of the response generation. One of completed, failed, in_progress, cancelled, queued, or incomplete.

status=(status : ResponseStatus | Nil)

The status of the response generation. One of completed, failed, in_progress, cancelled, queued, or incomplete.

usage

usage=(usage : ResponseUsage | Nil)