Which AI should students consider?
Maths, writing and research have different evidence. What it supports and where to double-check.
Facts below come from the datasets on this site, last updated 29 Oct 2026, 14:36 UTC. Interpretation is labelled as ours. We have not tested these products hands-on.
What the evidence shows
- On hard maths problems (FrontierMath Tiers 1-3) the top listed model solved 93.7%. GPT-6.1 Sol leads.
- On short factual questions the top score is 76.0%. Even the leader gets a substantial share wrong on that test.
- Writing is judged by Arena voters, who show preference. That does not show whether an essay's claims are right.
Our interpretation
An AI can be useful for explaining a method or testing your understanding. Because even the best listed score on factual questions is well below 100%, anything you will rely on needs checking against a reliable source. Follow your school's rules on AI use.