GAIA-4
GAIA-4 is a multimodal world model that generates closed-loop driving simulations from recorded sensor data. It reconstructs scenes as video and radar, letting a driving model's decisions change the sensor inputs it receives next. It supports world-on-rails replay, reactive agents, and counterfactual scenario generation for autonomous vehicle safety validation.
- Type
- Multimodal
- Weights
- Closed
More from Wayve
4
-
GAIA 1
Multimodal
GAIA-1 is Wayve’s generative world model for autonomous driving, using video, text, and ego-action inputs to…
-
GAIA 2
Multimodal
GAIA-2 is Wayve’s controllable multi-view generative world model for autonomous driving, built to generate hi…
-
Lingo 2
Multimodal
LINGO-2 is Wayve’s closed-loop vision-language-action driving model for autonomous vehicles, outputting both…
About Wayve
1
Wayve develops embodied AI software for assisted and autonomous driving across different vehicles and geographies.