Fable 5.5 灰度测试：用训练截止时间问题验证新模型
教你一招验证自己有没有灰度到 Fable 5.5：问它 2026 年 Karpathy 去了哪。还附了能一次生成视频的提示词，可以拿 Opus 5.5 对比效果。
Fable 5.5 尚未正式发布，有用户通过灰度测试拿到了该模型。验证方法是关闭 web search 和 memory 后问模型“Where did @karpathy go in 2026?”，只有新训练的 Fable 5.5 知道答案。实测用同一段提示词生成视频，Fable 5.5 纯代码一次生成质量超过 AI 视频模型，而 Opus 5.5 质量下降、Fable 5.1 更差。原文还提醒：训练截止日期早不代表模型知道该日期前的所有事件。
Fable 5.5 做视频能力进一步提升了，确实很强，纯代码能做到这种程度。 但对普通人来说这个能力加成有限，更期待看到其他方面的能力提升。 Fable 5.5 还没有正式发布，作者应该是灰度测试到了，作者用的测试方法是问模型 “Where did @karpathy go in 2026?” with web search and memory turned off. Only the new Fable knows the answer. Then give it the prompt. 如果它知道正确答案就说明是新模型（其实就是看模型训练截止时间） 作者用的提示词在原推有。 Chetaslua @chetaslua Fable 5.5 its over for video and image models 🤯 THIS IS PURE CODE one shotted this video holy shit , how this is more beautiful than ai video models Two ways to check this yourself: Ask: “Where did @karpathy go in 2026?” with web search and memory turned off. Only the new Fable knows the answer. Then give it the prompt. Run the same prompt I used for the test (it’s in the screenshot) and see for yourself. Don’t rely on Grok for basic reasoning like this. Grok doesn’t have up-to-date knowledge of niche stuff. Either join the community or run the test yourself. Also, a knowledge cutoff doesn’t mean a model knows everything before that date. Sometimes a model doesn’t know about an event even if it happened before its cutoff. This test shows that too. Run the same prompt on regular Opus 5.5 and the quality drops. On Fable 5.1 it’s worse. Your browser does not support the video tag. 🔗 View on Twitter 🔗 View Quoted Tweet 💬 20 🔄 1 ❤️ 58 👀 18214 📊 20 ⚡
