Our unique approach partitions physical GPUs into discrete virtual GPUs, each with their own memory and compute. Perfect for AI inference and model building workloads such as natural language processing, voice recognition and computer vision.
Vultr offers virtual machines powered by variety of graphics cards, so you can accelerate a wide range of workloads.
Spin up a new instance with a set of GPU-ready images that pre-install NVIDIA drivers, popular AI/ML frameworks, or game servers.