Zero-shot video (or image-frame) → code results on the test set, across commercial and open-source models.
Each output is tagged with A = appearance similarity and T = temporal similarity; higher is better for both. Click a video to inspect its code.
49–56 of 214
ground truth
Only CSS: Milk
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout
ground truth
Only CSS: Kirby
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout
ground truth
Only CSS: Bicycle
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout
ground truth
Only CSS: Screw 2
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout
ground truth
Only CSS: Grid Template Animation
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout
ground truth
Only CSS: Japanese "人力車"
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout
ground truth
Only CSS: Porco Rosso "SAVOIA" (紅の豚)
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout
ground truth
Only CSS: Lanthanum
model outputs
Gemini 3 Flash Preview
Qwen3-VL-8B-Instruct
GPT-5.4
Claude Sonnet 4.6
LLaMA 4 Scout