**Anthropic Opus 5在ARC-AGI-3基准测试中大幅领先GPT-5.6 Sol**

_Anthropic&\#039;s Opus 5 blows past Fable 5 and GPT-5.6 Sol on the benchmark designed to measure real intelligence_

> Anthropic的Claude Opus 5在旨在衡量真实智能的ARC-AGI-3基准测试中得分30.2%,远超GPT-5.6 Sol的7.8%。开发者指出,该模型展现出前所未有的独立反思方程生成能力,体现了更强的逻辑推理水平。

**来源信息**
- **来源**:The Decoder:AI News(RSS)
- **分类**:ai-models
- **发布时间**:2026-07-26 17:43(北京时间)
- **原文**:[打开原文](https://the-decoder.com/anthropics-opus-5-blows-past-fable-5-and-gpt-5-6-sol-on-the-benchmark-designed-to-measure-real-intelligence)