Anthropic 称 Claude Fable 5.2 完成三项硬核任务
这条推文说 Claude Fable 5.2 一个 prompt 做出三件能用的电子设备，还解了 72 个开放优化问题。硬件和化学方向的人可以关注，但都是 Anthropic 自述，等官方报告或第三方复现再下结论。
Anthropic 透露，Claude Fable 5.2 仅凭一段简短提示词就设计出三件可正常工作的电子设备。
在另外两组独立测试中，该模型还提出了未经实验验证的化学合成路线，并对 72 个公开收录的开放优化问题给出了优于已有列出的解法的方案。这些结果来自 Anthropic 自己的测试描述，目前尚未看到第三方复现或完整技术报告。
holy moly Anthropic says Claude Fable 5.2 designed three working electronic devices from a short prompt.
In separate tests, Claude also proposed untested chemical synthesis routes and improved on listed solutions to 72 open optimization problems.
Yeah, 2026 ends with a bang. And gives us a glimpse of all the crazy things that'll happen in 2027.
This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Claude Fable 5.2, Anthropic, 推理模型, 优化问题.
Source attribution: curated from kimmonismus (@kimmonismus).
