Claude Opus 4, Claude Sonnet 4
Anthropic
partial detail 19 method fields published
Evaluation
Internal AI research evaluation suite 2
ability to autonomously perform self-contained AI/ML research tasks across subareas relevant to Anthropic research (AI R&D-4 RSP threshold)
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.