METR finds Claude Opus 4.5 achieves record 4-hour 49-minute time horizon on AI autonomy tasks, highest score published to date

METR on X: "We estimate that, on our tasks, Claude Opus 4.5 has a 50%-time horizon of around 4 hrs 49 mins (95% confidence interval of 1 hr 49 mins to 20 hrs 25 mins). While we're still working through evaluations for other recent models, this is our highest published time horizon to date. https://t.co/N2fyrdulvR" / X

PostConversationMETR@METR_EvalsWe estimate that, on our tasks, Claude Opus 4.5 has a 50%-time horizon of around 4 hrs 49 mins (95% confidence interval of 1 hr 49 mins to 20 hrs 25 mins). While we're still working through evaluations for other recent models, this is our highest published time horizon to date.8:25 PM · Dec 19, 2025667.4K Views...