Costs are API-equivalent: computed per request from official prices, even though the runs use subscriptions. Scores are automatic checks; nothing is graded by hand unless marked.
Anthropic models run in Claude Code 2.1.296 · OpenAI models run in Codex CLI 0.162.0 · no harness of our own (yet) · why it matters
Data as of Oct 10, 20:15 UTC · How we test · Models · Every run
