Ask HN: What are the things LLM models fail at?

By aqua_coder · 2026-10-19 · 1 points · 0 comments

Generally, whenever I do see frontier AI model's evaluation metrics on say Artificial Analysis. It gives out a benchmark on a set of questions (or tasks). But it never quite actually shows what tasks it fails at. I'm more curious on benchmarks that focus on SWE.

Open the full discussion on BetterNews