Skip to content

Dataset · figma-ai

Motion graphics generation benchmark from Figma and Stanford

figma-ai/motion-graphics-bench

Expert benchmark with programmatic verification to evaluate how LLMs animate vector graphics.

What's inside

The joint benchmark from Figma and Stanford University researchers includes text, images, and video (10 to 100 thousand items). It was created to test the ability of large language models (LLMs) to turn static vector graphics into animation using programmatic verification.

What you can do with it

  • Evaluating and comparing LLMs in vector graphics animation code generation tasks
  • Developing and testing AI assistants for motion design
Size
10K–100K rows
Downloads this month
948
Likes
11

How to load it

from datasets import load_dataset

ds = load_dataset("figma-ai/motion-graphics-bench")