METR's Predeployment Evaluation of Claude Opus 5.5
What happened
METR 在模型发布前测了 Claude Opus 5.5 对 AI 研发的影响。结论是它比 Fable 5.1 强一点，在预算版 NanoGPT 速通赛、游戏机器人这类可验证任务，以及 LMCA 概念推理、Sunlight 开放研究这类难验证任务上都有提升，但不是断崖式飞跃。Anthropic 内部问卷也说它延续了 Mythos 级别的趋势。模型在需...
Coverage
Follow the reports to see the story from different sides.
- AI HOT (Curated Pool)PickMETR's Predeployment Evaluation of Claude Opus 5.5
METR evaluated Claude Opus 5.5's impact on AI R&D. It's a modest step up from Fable 5.1, not a leap toward full automation. Gains showed on verifiable tasks like Budget NanoGPT and Gaming Bot, and on harder-to-verify ones like LMCA and Sunlight. Anthropic's internal questionnaire says it continues the Mythos-level trend. A separate, undisclosed METR report estimates AI already accelerated Anthropic's overall R&D by ~1.5x, with a 30% chance of 2x. The post doesn't disclose specific parameters, pricing, or a release timeline.