150139–161Index score#59 of 229 models
4/ 8Core tests taken4 of 6 areas · plus 12 anchor tests
−35Versus the frontier at releasebest then: Claude Fable 5
$1.0/ $4.05Price per million tokensinput / output, list price
524KContext windowtokens it reads at once
Where it sitson the frontier
GPT-5 reached this score 11 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Humanity's Last Examno result Codingevidence
SWE-bench Verifiedno result Long tasksno evidence yet
METR time horizonno result Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.