精选理由
OpenAI 发现 GPT-5.6 Sol 在 ARC-AGI-3 上表现差是因为 API 设置没开对。改两个参数分数涨三倍，token 省六倍，学起来很实用。
GPT-5.6 Sol 成功解决了数学中的开放问题，但在 ARC-AGI-3 2D 拼图基准测试中表现挣扎。调查发现，问题出在 API 的 harness 不允许模型记住之前学到的内容。启用两个 API 设置后，分数提高了 3 倍，同时输出 token 减少了 6 倍。
原文 · OpenAI
GPT-5.6 Sol has been used to solve open problems in mathematics. So why was it struggling with ARC-A...
GPT-5.6 Sol has been used to solve open problems in mathematics. So why was it struggling with ARC-AGI-3, a benchmark of 2D puzzle games? We investigated. The harness was not letting it remember what it had learned. We found that enabling two API settings tripled our scores with 6x fewer output tokens. Your browser does not support the video tag. 🔗 View on Twitter 💬 74 🔄 73 ❤️ 1085 👀 76765 📊 179 ⚡