**“万物计数”AI模型:用文本提示准确计数图像中对象,误差率减半**

_New AI model called "Count Anything" does exactly what it says, and that&\#039;s harder than it sounds_

> “Count Anything”是首个能通过文本提示对任何图像(如人群、显微细胞)中的物体进行计数的AI模型。在对比测试中,其误差率较以往系统降低一半,但在处理极高密度物体和模糊术语时仍有局限。

**来源信息**
- **来源**:The Decoder:AI News(RSS)
- **分类**:ai-models
- **发布时间**:2026-06-14 01:00(北京时间)
- **原文**:[打开原文](https://the-decoder.com/new-ai-model-called-count-anything-does-exactly-what-it-says-and-thats-harder-than-it-sounds)