Measuring AI Ability to Complete Long Tasks: Opus 4.5 has 50% horizon of 4h49M

via metr.org

Short excerpt below. Read at the original source.

Article URL: https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/ Comments URL: https://news.ycombinator.com/item?id=46342166 Points: 28 # Comments: 4

Read at Source