animation2code benchmark

animation2code benchmark

For best compatibility, please view this dashboard in a Chrome browser.

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.

209–214 of 214

ground truth

Animated Concepts #3

model outputs

Gemini 3 Flash Preview

A 0.77T 0.32

Qwen3-VL-8B-Instruct

A 0.63T 0.22

GPT-5.4

A 0.89T 0.22

Claude Sonnet 4.6

A 0.81T 0.27

LLaMA 4 Scout

A 0.44T 0.21

ground truth

Animated Concepts #3

model outputs

Gemini 3 Flash Preview

A 0.85T 0.27

Qwen3-VL-8B-Instruct

A 0.81T 0.19

GPT-5.4

A 0.91T 0.27

Claude Sonnet 4.6

A 0.78T 0.28

LLaMA 4 Scout

A 0.61T 0.22

ground truth

Animated Concepts #3

model outputs

Gemini 3 Flash Preview

A 0.84T 0.39

Qwen3-VL-8B-Instruct

A 0.80T 0.12

GPT-5.4

A 0.90T 0.28

Claude Sonnet 4.6

A 0.85T 0.25

LLaMA 4 Scout

A 0.82T 0.26

ground truth

Animated Concepts #3

model outputs

Gemini 3 Flash Preview

A 0.82T 0.32

Qwen3-VL-8B-Instruct

A 0.68T 0.20

GPT-5.4

A 0.88T 0.20

Claude Sonnet 4.6

A 0.82T 0.28

LLaMA 4 Scout

A 0.56T 0.32

ground truth

Animated Concepts #3

model outputs

Gemini 3 Flash Preview

A 0.84T 0.26

Qwen3-VL-8B-Instruct

A 0.82T 0.21

GPT-5.4

A 0.90T 0.28

Claude Sonnet 4.6

A 0.86T 0.26

LLaMA 4 Scout

A 0.70T 0.20

ground truth

Apple Loading Bar

model outputs

Gemini 3 Flash Preview

A 0.62T 0.23

Qwen3-VL-8B-Instruct

A 0.60T 0.27

GPT-5.4

A 0.63T 0.25

Claude Sonnet 4.6

A 0.70T 0.52

LLaMA 4 Scout

A 0.61T 0.22

← Previous27 / 27Next →