Natix teams with Valeo for multi-camera AI model

Natix Network and Valeo have announced a partnership to build an open-source multi-camera world foundation model, with the first iteration to launch using 5,000 hours of multi-camera driving data collected through Natix’s VX360 network across the United States, Europe and Asia. The model extends Valeo’s existing VaViM and VaVAM open-source frameworks, which were trained on front-facing camera footage, to incorporate 360-degree surround-view data for autonomous driving and robotics applications.

Natix said its decentralised camera network has collected over 80,000 hours of driving data in six months. World foundation models are designed to predict how real-world scenes evolve over time, enabling autonomous systems to train in simulated environments before physical deployment.

In a statement, Alireza Ghods, Chief Executive and Co-founder of Natix, said: “World foundation models are a once-in-a-generation opportunity—similar to the rise of large language models in 2017–2020. The teams that build the first scalable world models will define the foundation of the next AI wave: Physical AIs.”

The partners said subsequent stages will expand the dataset with additional driving data from new geographies, weather conditions and traffic patterns. Models, datasets and training tools will be released under an open-source framework for researchers and developers.

Source: Natix