Choosing a model
How to pick the right AI model for your study.
QuestionPunk offers three AI model tiers: Haiku (fast and economical), Sonnet (balanced), and Opus (deepest). This guide explains the trade-offs.
aimodels5-15 minutesIntermediateResearchersProduct teams
Steps
- Understand the optionsHaiku: Fastest responses, lowest cost. Good for simple follow-ups and high-volume studies.
Sonnet: Balanced speed and depth. Recommended for most research use-cases.
Opus: Richest follow-ups, highest cost. Best for deep qualitative interviews where nuance matters. - Select in survey settingsChoose the model in the AI interviewer settings for each survey. You can set different models for different interview questions within the same survey.
- Pilot and compareRun a short A/B pilot with different model settings to compare output quality before sending to your full sample.
Consider trade-offs between cost, speed, and depth. Haiku is cheapest and fastest; Opus provides the richest, most nuanced follow-ups for complex qualitative research.
Sonnet is the recommended default for most studies. It provides good follow-up depth at reasonable cost and latency.
Run a short pilot with 5-10 respondents using different model settings to compare output quality before committing to your full sample.