CoreWeave has signed a multi-year agreement to train Wayve’s AI Driver on its cloud as Wayve accelerates commercial deployment.
US-based CoreWeave is one of the so-called ‘neoclouds’ that provide the Graphics Processing Unit (GPU) compute, cloud infrastructure and software that businesses use to train and run large-scale AI models.
Wayve will train its foundation models on CoreWeave Cloud with NVIDIA Blackwell GPUs, expanding Wayve’s AI infrastructure for training, evaluation and simulation.
The Wayve AI Driver is Wayve’s self-driving software, which it develops for automakers and fleet owners. It uses end-to-end AI trained on real-world driving data from more than 70 countries and works without high-definition maps across a range of vehicles, from driver assistance systems to robotaxis.
Training a system like this takes cloud infrastructure that stays reliable through long training runs, with enough visibility for engineers to prove each new version is better and check its safety at scale, according to CoreWeave.
“AV developers need high-performance compute and the tools and support to use it effectively,” Dr Richard Ahlfeld, CoreWeave’s Senior Vice President for Physical AI, told IZON. “CoreWeave brings those capabilities together, helping teams run simulations, find edge cases and train and compare models faster. The technology is developing quickly, and our job is to help customers keep pace.”
CoreWeave said its cloud has posted record-breaking results in MLPerf, the industry’s standard benchmark for how fast AI systems train and run models.
The company is now putting more than 250,000 GPUs across 51 data centres and 1.5 gigawatts of active power behind the physical AI boom, as IZON revealed in an exclusive interview with its Physical AI leadership team in September.












