Question and answer

QuestionAnswerSample is the shape with no choices: a question and an answer.

Shape

from sdvm.types import QuestionAnswerSample

sample = QuestionAnswerSample(
    question="Natalia sold 48 clips in April and half as many in May. How many altogether?",
    answer="She sold 48 / 2 = 24 clips in May, so 48 + 24 = 72 altogether.",
)

On the wire:

{
  "task_type": "question_answer",
  "question": "Natalia sold 48 clips in April and half as many in May. How many altogether?",
  "answer": "She sold 48 / 2 = 24 clips in May, so 48 + 24 = 72 altogether."
}

answer is whatever the dataset publishes as the answer — a bare value, a sentence, or a worked solution. Nothing about the type is specific to one corpus. Both fields must be non-empty strings.

Extra columns

Anything beyond the two fields goes in extra and comes back untouched, exactly as on a multiple-choice item:

QuestionAnswerSample(
    question="...", answer="72",
    extra={"reasoning": "48 / 2 = 24, and 48 + 24 = 72.", "source_id": "gsm8k-train-0"},
)

That is also how a corpus that separates a reasoning trace from the value it arrives at is loaded: the value goes in answer, the trace rides along in extra and is returned as you sent it.

What the models return

sdvm-audit-1

What it returns for this sample.

sdvm-fix-1

What it changes, and what it never does.

Refine

Audit, fix and re-audit in one call.

Last modified on September 18, 2026

Get started

Sample types

Models

Guides

SDK

HTTP API

Pricing and limits

Question and answer

Shape

Extra columns

What the models return

sdvm-audit-1

sdvm-fix-1

Refine

​Shape

​Extra columns

​What the models return

sdvm-audit-1

sdvm-fix-1

Refine

Shape

Extra columns

What the models return