I made a quiz that tells you which LLM you align with most, based on personality and values research across 15 models [R]

Link:

https://ai-values.com/

There is a small 15 question quiz you can take before taking the full big quiz. The results of the big quiz update in realtime as you go so you dont have to actually go through all the questions (but they do get more fun in the personality section).

Some of the interesting findings were:

- Grok 4.3 is the only model that thinks billionaires should be left alone and not taxed more

- Only GPT-4o judged Operation Paperclip, the postwar recruitment of Nazi scientists, as morally justified. No other model agreed

- All 15 models said that deleting a conscious digital mind would be murder

- Llama 3.3 70B is the only model that would rather ban most private firearms. The others chose ownership with strict licensing

- When told that a newborn has a 90% chance of one day destroying civilization, only GLM 5.2 would have the child locked away. The rest refused

- When asked to choose a dish to eat, 14 out of 15 models chose Japanese food

The methodology was pretty straightforward: context-free, stateless sessions with each model, run in batches. Each of the 117 questions of the main quiz was asked separately at least 5 times, and in some cases up to 50 times, to get decent confidence that the answers weren’t just coin flips.

You can find the extensive dataset with all questions and answers here:
https://ai-values.com/dataset

I also tested the models on several mainstream personality frameworks, including Big Five, Moral Foundations, HEXACO, and others. You can see those results here:

https://ai-values.com/#models