Svelte Hacker News logo
  • top
  • new
  • show
  • ask
  • jobs
  • about

Not Even Bronze: Evaluating LLMs on 2025 International Math Olympiad

matharena.ai

3 points by amichail 9 hours ago

davydm 7 hours ago

No surprises. Math requires understanding, not rote autocompletion. LLMs are not suited to this task, or any requiring consistent precision.

  • asey 5 hours ago

    Is that so? https://x.com/gdb/status/1946479692485431465?s=46