Claude Opus 4.8 | Humanity's Last Exam Full set — No tools

Anthropic's Claude Opus 4.8 evaluation results on Humanity's Last Exam Full set — No tools

Score: 49.8%

Reasoning level: adaptive thinking, max

Source

Claude Opus 4.8 (Registry No. 20011)

Anthropic

Humanity's Last Exam Full set — No tools