Today for AI
HOT RADAR
llmHEAT 12.9°

AI CURATED HOT STORY · 10/26/2026

Code2Video-AA Benchmark Launches to Evaluate AI Video Generation via Code

1 reports archived1 independent sourcesupdated 10/26/2026, 08:00:00
Synthesis & Latest Updates
1 Sources Cross-Validated

Artificial Analysis launched Code2Video-AA, an independent implementation of HeyGen's Code2Video Bench, evaluating how well LLMs design short videos using HyperFrames HTML code. Claude Opus 5.5 leads with a significant Elo score advantage, ranking first across all five judge axes. This benchmark provides developers with quantitative insights into model performance in creative coding and multimodal output scenarios.

LATEST/Artificial Analysis launched Code2Video-AA, an independent implementation of HeyGen's Code2Video Bench, evaluating how well LLMs design short videos using HyperFrames HTML code. Claude Opus 5.5 leads with a significant Elo score advantage, ranking first across all five judge axes. This benchmark provides developers with quantitative insights into model performance in creative coding and multimodal output scenarios.

Hourly heat curve

2 fully observed hours
Fewer than 3 fully observed hours; trend not plotted yet.

Coverage timeline

Total 1 reports · Latest first
  1. Artificial AnalysisT1·68 pts

    Code2Video-AA Benchmark Launches to Evaluate AI Video Generation via Code

    • Code2Video-AA is Artificial Analysis' independent implementation of HeyGen's Code2Video Bench, focusing on models' ability to generate product/brand videos via code (HyperFrames HTML).
    • Claude Opus 5.5 leads the initial leaderboard with an Elo score of 1536, surpassing the second-place model by 93 points and ranking first across all five judge axes.
    • The benchmark evaluates 114 design briefs to assess code generation quality and visual presentation effectiveness in realistic creative tasks.