Simon Willison(@simonw)
I tried a 15MB, 30 page text-heavy PDF and Opus 4.7 reported 60,934 tokens while 4.6 reported 56,482...
5.0内容质量
TL;DR · AI 摘要
Simon Willison测试显示Opus 4.7对PDF的token计数比4.6高8%，但增幅远低于纯文本，暗示模型对结构化文档处理更稳定。
核心要点
- Opus 4.7相比4.6对PDF的token计数仅增加8%，增幅温和。
- 结构化文档（如PDF）的token膨胀率显著低于原始文本。
- 版本升级对不同输入格式的影响差异值得关注。
#Opus#Token计数#AI模型
打开原文Simon Willison on X: "I tried a 15MB, 30 page text-heavy PDF and Opus 4.7 reported 60,934 tokens while 4.6 reported 56,482 - that's a 1.08x multiplier, significantly lower than the multiplier I got for raw text." / X
Don’t miss what’s happening
I tried a 15MB, 30 page text-heavy PDF and Opus 4.7 reported 60,934 tokens while 4.6 reported 56,482 - that's a 1.08x multiplier, significantly lower than the multiplier I got for raw text.
·
3
14