Artificial Analysis 完整文章·· 2 小時前AI 評分38
Code2Video-AA：AI 模型以程式碼設計影片的表現如何？
Code2Video-AA: how well do AI models design videos in code?
AI 導讀
Code2Video-AA 評測 AI 模型以程式碼設計影片的表現，Claude Opus 5.5 在全部五個評分軸領先。每項任務的成本相差超過 300 倍，而所有模型至少成功算繪 96% 的影片。
正文
Overview
Example tasks
Three tasks from the public split of Code2Video Bench, each with every model's rendered video. Models are in order of Code2Video-AA Elo.
How Code2Video-AA works
Claude Opus 5.5 leads on all five axes
Cost per task varies more than 300x
Every model renders at least 96% of its videos
Code2Video-AA resources
來源：Artificial Analysis 完整文章 · artificialanalysis.ai