Source: Arm Author: Arm
    Arm Holdings Limited (NASDAQ: ARM, hereinafter referred to as "Arm") today announced the acquisition of the Arm? Kleidi technology extends to the automotive market. Arm Kleidi is a broad software and software community engagement program designed to accelerate the development of artificial intelligence (AI). With Arm Kleidi, partners will be able to fully exploit the benefits of next-generation application-specific AI models in vehicles and accelerate their deployment.
    Suraj Gajendra, vice president of products and solutions at Arm's Automotive division, said: "In the automotive industry, AI is often referred to as the heart of autonomous vehicles - which sounds futuristic and is still years away from being ready for mass production. However, the reality is that many applications and functions in today's new vehicles, from adaptive cruise control, personalized in-vehicle infotainment systems to driver and passenger monitoring systems, are already using AI technology, and these application-specific models and workloads are already running efficiently on the Arm platform. The flexibility, performance, power efficiency and scalability of the Arm computing platform make it the platform of choice for developers across the automotive industry."
    However, as these automotive applications and functions continue to evolve, the demand for more AI capabilities and increasingly complex workloads continues to rise, and the automotive industry is facing a unique set of challenges, including:
Compared to other industries, the automotive industry has a large and complex software system, which makes the development environment extremely complex.
    A robust cloud-to-vehicle software development and verification infrastructure is needed to seamlessly deploy cloud-developed applications into the car.
    The software must meet stringent industry standards and regulations for automotive functional safety and information security.
    Cars have a long life cycle, which means software must be upgraded to accommodate new features and updates in the future (similar to synchronizing updates to smartphones when the latest software version is released, but cars have a longer product cycle).
    Arm addresses these unique challenges through leading Automotive Enhancement (AE) technology and collaboration in the automotive ecosystem. These collaborations cover virtual platforms and standards-based software solutions to accelerate the automotive development cycle through scalable Open Architecture (SOAFEE) for embedded edge sides. All of these solutions provide developers with the tools to seamlessly achieve higher performance through the widely used Arm computing platform and accelerate time to market for their products.

Automatic performance optimization with Arm Kleidi
    Launched in 2024, Arm Kleidi is designed to provide software performance optimization for AI inference workloads running on Arm cpus without any additional work by developers, and the software library is now used in key areas such as mobile, cloud, and data centers. Including Kleidi, which has been integrated into the latest versions of popular AI frameworks such as ExecuTorch, Llama.cpp, MediaPipe, and PyTorch, developers just need to start building applications to automatically capture significant performance gains on ARM-based platforms.
    Kleidi's performance optimization features have been introduced to the automotive market, with leading technology companies already experiencing performance enhancements for a range of new applications in their automotive software. From in-car chatbots and personalized driver suggestions, to image and motion enhancements to assist users, diagnose and solve problems, the integration of Kleidi libraries with key developer frameworks is accelerating the development of AI applications to provide users with faster and more efficient in-car experiences.

AWS Automotive has increased the responsiveness of in-car chatbots by 10 times
    Using a virtual platform jointly developed by Arm and Corellium, Amazon Cloud Technologies (AWS) has created a new proof-of-concept AI chatbot for software-defined vehicles (SDVS) that allows drivers to interact directly with the car. Ask questions or ask for dashboard alerts without leafing through the car's owner's manual. With the latest version of Llama.cpp developed chatbot integrated with KleidiAI, AWS has achieved a significant performance improvement in its chatbot, with response times reduced to one to three seconds and performance improved by a factor of 10. KleidiAI saved six weeks of chatbot development time because developers didn't have to spend much effort on underlying software optimization.

VicOne accelerates its response to automotive cybersecurity threats

    VicOne's xCarbon in-vehicle solution for SDVS enables vehicles to learn and identify cybersecurity threats, reducing their reliance on cloud technology, thereby reducing costs and better securing data. With the TinyLlama-11B model, the VicOne solution has achieved significant performance improvements, doubling the pre-fill prompt speed and increasing lexical generation by more than 60%, thereby reducing vehicle response time to detect cybersecurity threats. The performance optimizations implemented by Kleidi enable VicOne to deliver a safer and more reliable driving experience to users without the need for additional development work.
    Sonatus' AI Technician Builder platform, built for Oems, runs seamlessly on ARM-based AWS Graviton cpus
    Sonatus' AI Technician Builder is a cloud-to-edge platform that helps Oems create customer service agents that respond quickly to vehicle queries to improve the driving experience. The platform initially runs on Gpus in cloud environments, but with the introduction of Kleidi optimization framework, it can now be easily ported to run on AWS Graviton EC2 cloud instances with Arm architecture. This not only ensures the cost performance and response time required by users, but also paves the way for running edge AI workloads on automotive computing platforms based on Arm architecture.

Driving AI innovation in automotive and beyond on the Arm platform
    Today, 94% of the world's automakers use Arm technology in their latest models. The widespread use of the Arm computing platform in the automotive sector shows that it is uniquely positioned to drive significant performance optimizations for critical in-vehicle AI applications, as exemplified by AWS, VicOne, and Sonatus. Integrating Kleidi into existing software stacks enables more efficient and advanced AI use cases and provides faster response for critical automotive applications where every minute counts. This gives developers the power to optimize and run new application-specific AI models and workloads on Arm cpus without additional cost or development effort.
    By expanding Kleidi into automotive, Arm continues to deliver on its promise to deliver seamless performance acceleration for AI developers across Arm application markets, from the cloud to the edge, enabling them to create compelling new user experiences and unlock the transformative potential of AI.

About Arm
    As the industry's most powerful and energy-efficient computing platform, Arm covers 100% of the world's connected population with unparalleled scale. Arm delivers advanced solutions to meet the never-ending demand for computing, enabling the world's leading technology companies to unlock unprecedented AI experiences and performance. Arm is working with the world's most extensive computing ecosystem and 20 million software developers to build the future of AI on the Arm platform.