Skip to content

GenAD

Generate driving videos for autonomous driving research

Visit official site

What GenAD is

The model is described as a large-scale video generation model for autonomous driving that predicts and simulates real world driving scenes for research and application.

What you can do with it

  • Predict and simulate real driving scenes
  • Control generated video with language and trajectories
  • Apply the output to driving planning tasks

Application scenarios

  • Researchers generate complex driving videos to train and test algorithms
  • Driving companies produce large volumes of labelled data
  • Simulation environments gain realistic and dynamic virtual scenes

Pros & cons

✓ What we like

  • Handles complex dynamic environments and open-world scenes
  • Video output can support training and testing

! What to watch out for

  • Control relies on language and trajectory inputs
  • The planning use is described as a potential application

FAQ

What does it generate?

Videos that predict and simulate real driving scenes.

How is generation controlled?

Through language and driving trajectories.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review