Skip to content
Inside the Frontier

Evaluation

Internal AI research evaluation suite 1

ability to improve performance of machine-learning code across LLMs, time series, low-level optimizations, reinforcement learning, and general problem solving

Autonomy & self-improvement Anthropic
Ask about this evaluation

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.

Claude Opus 4, Claude Sonnet 4
Anthropic

rich detail 24 method fields published

See how this test was run