Making Arm KleidiAI Work for Your ML Framework

Explore how KleidiAI, an optimization library designed to accelerate Machine Learning workloads on Arm CPUs, can be integrated into various ML frameworks in this 24-minute conference talk. Learn about the library's "take-what-you-want" philosophy and how it has been rapidly adopted since its launch in May 2024. Discover different integration styles, implementation strategies, and methods to achieve comprehensive workload acceleration coverage. This presentation provides valuable insights for developers looking to optimize ML performance on Arm architecture through effective KleidiAI integration techniques.