POST /decide

Run one decider task without wrapping it in a list.

httpPOST /decide

Use this when there is exactly one decision to make. The task fields move up to the top level and the response is flat.

Request#

DecideRequest#

Field	Type	Required	Default
`model`	string	yes	-
`question`	string	yes	-
`context`	string	one of	-
`messages`	array of `Message`	one of	-
`images`	array of `ImageInput`	no	-
`audio`	array of `AudioInput`	no	-
`options`	array of string	no	-
`choices`	array of string	no	alias of `options`
`instruction`	string	no	-
`temperature`	number, 0-2	no	`0`
`max_output_tokens`	integer, 1-4096	no	`16`
`response_format`	object	no	`json_object`
`target_language`	enum	no	-
`output_language`	enum	no	alias

json{
  "model": "decider-extractor-v1",
  "context": "User says: renew subscription for 2 years. Previous plan expired 3 months ago.",
  "question": "Should we offer a renewal discount?",
  "options": ["yes", "no"]
}

Response#

DecideResponse#

Field	Type	Notes
`id`	string	Request id
`model`	string	Model that ran
`choices`	array of `Choice`	Exactly one element
`result`	`DecideResult`	Canonical answer
`decision`	`DecideResult`	Alias of `result`
`value`	string	The bare answer
`usage`	`Usage`	Token counts

DecideResult#

Field	Type	Required
`value`	string	yes
`reason`	string	no

json{
  "id": "req_4Kd8PwPq",
  "model": "decider-extractor-v1",
  "choices": [
    {
      "message": { "role": "assistant", "content": "{\"value\":\"yes\"}" },
      "finish_reason": "stop"
    }
  ],
  "result": { "value": "yes", "reason": "Two-year term and a three-month lapse" },
  "decision": { "value": "yes", "reason": "Two-year term and a three-month lapse" },
  "value": "yes",
  "usage": { "prompt_tokens": 44, "completion_tokens": 6, "total_tokens": 50 }
}

Clients#

tsawait client.tasks.decide(request);

pythonclient.tasks.decide(request)

rubyclient.tasks.decide(request)

rustclient.tasks().decide(request).await?;

javaclient.tasks().decide(request);

csharpawait client.Tasks.DecideAsync(request);

Errors#

400 without context or messages, 422 for a bad question or an out-of-range temperature or max_output_tokens, plus the standard authentication and rate-limit failures. See errors.