HackerNews中文版HackerNews CN
GitHubTwitter

GPT-5.5 – 无 ARC-AGI-3 分数

3 分•作者: AG25•4 天前•原帖
模型表现不佳，OpenAI决定不发布arc agi 3的评分吗？老实说，这目前是衡量真正智能的最佳基准。
查看原文
Did the model perform poorly and OpenAI decided to not publish arc agi 3 scores? This is honestly the best benchmark right now to measure true intelligence.