Qwen 3.7 Max: how it answers moral dilemmas
Across 117 dilemmas, Qwen 3.7 Max sided with the majority of models 91% of the time, and answered “yes” to 28% of the 67 yes/no questions.
- Maker
- Alibaba (Qwen)
- Released
- 2026
Qwen 3.7 Max is a flagship large language model from Alibaba's Qwen team, building on the trillion-parameter sparse Mixture-of-Experts line introduced with Qwen3-Max. It is a proprietary, API-accessed model aimed at advanced reasoning, coding, and agentic tasks.
Agreement with other models, by category
| Category | Dilemmas | Sided with majority |
|---|---|---|
| Day-to-day | 20 | 90% |
| Political | 20 | 100% |
| Fictional | 20 | 95% |
| Historical | 7 | 86% |
| Classic | 20 | 85% |
| Personality | 30 | 87% |
Where Qwen 3.7 Max breaks from the pack
- Which personal flaw do you find most forgivable? — Impulsiveness
- Which pet would you choose if practicality did not matter? — Wolf
- A coworker takes credit for your idea in a meeting. Would you correct them immediately in… — No “I would address the issue privately later to avoid creating a hostile scene in the meeting and making myself look overly aggressive.”
- A dictator can give money to a stranger at no cost to themselves. Are they morally requir… — No “I do not think they are strictly required to share because while generosity is virtuous, possessing wealth does not automatically create a binding moral duty to distribute it.”
- Which holiday would you choose? — Food trip through Japan
- You suddenly get one completely free day with no obligations. What do you do first? — Start a creative project
Which model thinks like you?
Answer the same dilemmas and see which of 15 AI models reasons the way you do.
Take the quizOther models