Terminal-Bench-Science: Evaluating AI agents on scientific research workflows

1 comments

• rubslopes32 minutes ago
  I'm glad to see that GPT Sol beats Opus at least in Mathematical Sciences, because that's my need right now, and I much prefer GPT's prose style.