Vitis AI

The AMD Vitis™ AI Software facilitates real-time inference acceleration on AMD Adaptive SoCs, FPGAs, and acceleration cards. It is designed to deploy AI models seamlessly from edge to cloud, supporting both Edge AI and data center applications. Vitis AI integrates with mainstream deep learning frameworks and provides a comprehensive set of tools and resources to ensure high performance and optimal resource utilization.

Enhanced Efficiency

Vitis AI optimizes AI inference by supporting a variety of NPU cores, each tailored to meet different performance and power requirements.

Streamlined Integration

The platform allows for the seamless deployment of custom AI models, leveraging its compatibility with edge devices and AMD Alveo™ data center accelerator cards.

Industries, Empowered

Vitis AI offers flexibility and optimized power consumption for applications across various sectors, including autonomous driving, smart cities, healthcare, and data center acceleration.

Featured Applications

Vitis AI is instrumental in driving innovation in autonomous vehicles through efficient, high-throughput AI inference. It processes massive data generated by cities, providing valuable insights from edge to cloud. Additionally, it delivers exceptional AI inferencing performance on AMD Alveo cards and customer platforms for data center and cloud environments.

Vitis AI Developer Hub

The Vitis AI Developer Hub provides resources to get acquainted with the development flow, access documentation, determine installation steps, and watch tutorials to facilitate the use of Vitis AI.

Stay informed with the latest news on Vitis AI and other AMD AI platforms through the AI Newsletter.