Vchitect/VBench

[CVPR2024 Highlight] VBench - We Evaluate Video Generation

68
/ 100
Established

This project helps video creators and evaluators rigorously assess the quality and capabilities of AI-generated videos. You provide a video generation model, and it outputs a detailed report on its performance across various dimensions like motion, aesthetics, and consistency. This is ideal for researchers, AI developers, and quality assurance specialists working with generative video AI.

1,537 stars. Actively maintained with 4 commits in the last 30 days. Available on PyPI.

Use this if you need to objectively benchmark and understand the strengths and weaknesses of different video generative AI models or evaluate your own model's output comprehensively.

Not ideal if you are looking for a simple tool to just generate videos or need a basic pass/fail quality check without detailed analytical breakdowns.

video-generation AI-evaluation generative-AI computer-vision AI-benchmarking
Maintenance 16 / 25
Adoption 10 / 25
Maturity 25 / 25
Community 17 / 25

How are scores calculated?

Stars

1,537

Forks

107

Language

Python

License

Apache-2.0

Last pushed

Mar 16, 2026

Commits (30d)

4

Dependencies

25

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Vchitect/VBench"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.