On June 4 local time, semiconductor IP giant Arm officially launched its new automotive product brand, Zena Compute Subsystem (CSS), to accelerate the development of next-generation AI chips for automotive design.
Zena CSS integrates 16 Armv9-based Cortex-A720AE cores andCortex-R82AE real-time microcontroller cores, and is optimized for performance in ADAS and IVI applications. It also features optional image signal processing supported by Mali-C720AE and Mali GPUs.
Today, nearly all global OEMs, including Tesla, Rivian, NIO, Mercedes-Benz, Honda, and Geely, rely on Arm technology as the foundational computing platform to support everything from safety-critical ADAS to immersive in-car experiences. The next generation of vehicles will require more: ubiquitous personalized intelligence that can adapt and update in real-time throughout the vehicle.
To compete in the AI era, automakers need scalable computing without compromising safety, energy efficiency, or flexibility.
Arm states that Zena CSS’s standardized pre-integrated computing platform can significantly shorten development cycles and enable software teams to start early, even before the chips are available. This accelerates time-to-market, expands AI capabilities across various vehicle models, and helps automakers remain competitive in the rapidly evolving AI-defined automotive landscape.
Specifically, Zena CSS reduces chip development time by up to 12 months and decreases the chip engineering workload for each project by up to 20%, enabling OEMs and chip suppliers to bring new models to market a year earlier through smart voice and touch interfaces, immersive digital cockpit experiences, driver assistance and automation, and real-time perception.
Overview of Arm Zena CSS
The next generation of vehicles needs to run various workloads on a single platform, including Advanced Driver Assistance Systems (ADAS), In-Vehicle Infotainment (IVI), and vehicle control systems, each with different safety requirements and real-time constraints. Zena CSS is built on Armv9 Automotive Enhanced (AE) technology and is our first pre-integrated and pre-validated platform for AI-defined vehicles, offering:
16 Armv9-based Cortex-A720AE cores, optimized for ADAS and IVI applications;
A safety island driven by Cortex-R82AE, unique to Zena CSS, for real-time ASIL D processing capabilities such as fault management, safety monitoring, system control, and SoC boot;
A runtime security engine with a secure hardware root of trust (RoT), leveraging Arm TrustZone technology to manage SoC-level security;
CPU consistency and chip-to-chip connectivity provided by CMN S3AE;
Optional image signal processing supported by Mali-C720AE and Mali GPUs, suitable for ADAS use cases, including surround view and driver monitoring;
Support for easy integration of accelerators and partner-specific logic to meet the evolving workload demands of AI-supported advanced SoC designs;
This standardized computing platform enables OEMs and chip providers to reuse architectures and software, accelerating global vehicle delivery, simplifying system development and integration, while streamlining functional safety and information security certification. This is crucial, as developing customized hardware and software stacks for each vehicle model and application is costly, time-consuming, and fraught with validation and certification challenges. Importantly, Zena CSS allows our partners to innovate and differentiate on a standard platform, adding unique value to their product lines.
Leading OEMs (including top electric vehicle manufacturers) and major chip suppliers have already licensed Zena CSS or are in advanced stages of participation. Arm expects that in the coming years, most of the industry will be based on the Zena CSS series.
Software innovation time-to-market reduced by 2 years
In addition to shortening chip time-to-market, Arm’s partners can leverage cloud-based virtual platform support from AWS, Cadence, Siemens, and Synopsys to immediately begin software development on Zena CSS, thereby reducing software innovation time-to-market by up to two years.
Arm states that the developer ecosystem consists of over 20 million developers (including Denso, GitHub, Green Hills, Mapbox, Red Hat, Panasonic Automotive, etc.) who can now build and test across the entire software stack and general computing platform. Zena CSS is built on open standards, including AUTOSAR, COVESA, eSync, and Virtio, and is supported by SOAFEE, which has over 150 members, including General Motors, CARIAD, Continental, and Tata Motors.
The AI-defined automotive era on Arm
In the past five years, the shipment of Arm-based chips in the automotive market has doubled, demonstrating the industry’s trust in our platform. Now, with Zena CSS, we provide automakers with the tools to build software-defined, AI-supported vehicles faster, smarter, and more efficiently. Combined with the unique advantage of cloud-to-automobile parity, where development and testing occur on the same Arm architecture in the cloud that will be deployed in the vehicle, our partners can be confident that software will run seamlessly from minimal sensors to Arm-based clouds.
By shortening chip development time by 12 months, reducing engineering workload, and supporting earlier software innovation, Zena CSS helps automakers deliver AI-defined experiences at least a full model year ahead—all achieved on Arm.
Partner Quotes:
Keiichi Aida, Chief Technology Officer and Chief Information Security Officer at Astemo, Ltd. (formerly Hitachi Astemo), stated, “Astemo has been closely collaborating with Arm to realize the advantages of SOAFEE through our Internet of Vehicles (IoV) platform. With the launch of Arm Zena Compute Subsystem (CSS), we see an opportunity to further accelerate development by porting our IoV platform to a standardized computing architecture. Arm Zena CSS enables us to scale more efficiently, leveraging the advantages of virtual prototyping and meeting the growing demands of software-defined vehicles. We are excited to continue our collaboration with Arm and help drive the industry towards a more connected and software-centric future.”
Dr. Martin Tietze, Vice President of New Technologies at BMW Group, stated, “At BMW, we continuously explore innovative approaches to future automotive computing architectures. Currently, we focus on the co-design of hardware and software for advanced embedded machine learning applications, such as autonomous driving features, and are committed to defining future chip reference architectures for an open automotive chiplet ecosystem. Arm Zena Compute Subsystem (CSS) provides an excellent starting point for our semiconductor simulation work. Arm’s architecture is deeply embedded in the automotive ecosystem, and we greatly welcome and appreciate Arm’s support and collaboration in programs like imec’s Automotive Chiplet Program.”
Magnus Östberg, Chief Software Officer at Mercedes-Benz, stated, “The transition to software-defined vehicles requires us to change the way we handle computing architectures. Standardized, pre-validated computing subsystems like Arm Zena CSS can significantly accelerate development time and reduce complexity across the industry. By integrating such solutions early in the development chain, the automotive industry can simplify complexity and lay a stronger foundation for delivering safe, intelligent, and connected experiences that define the next generation of vehicles. At Mercedes-Benz, we recognize the value of collaborative platforms that drive the entire ecosystem forward.”
Masashige Mizuyama, Executive Vice President and Chief Technology Officer at Panasonic Automotive Systems Co., Ltd., stated, “At Panasonic Automotive Systems, we are committed to accelerating the development of software-defined vehicles, and Arm Zena Compute Subsystem represents an important step forward in this transformation. Based on our ongoing collaboration with Arm and our work, including within the SOAFEE community, we see tremendous value in porting the SOAFEE blueprint based on the standard device virtualization framework VirtIO (e.g., unified HMI and vSkipGen) to the standardized, scalable platform provided by Arm Zena CSS. Virtual prototyping based on a general computing foundation not only shortens time-to-market but aligns with our vision of building a more software-centric and flexible automotive ecosystem. With Arm’s widespread presence across the industry, we are excited to deepen our collaboration and help shape the future of automotive computing.”
Jang-kyu Lee, CEO of Telechips Inc., stated, “Telechips has been a long-term partner of Arm, and this ongoing collaboration is key to our mission of providing high-performance, cost-effective semiconductor solutions for the global automotive market while shortening development times. Arm Zena Compute Subsystem (CSS) provides a standardized and scalable computing platform that enables us to accelerate development and focus on unique areas of innovation. Arm Zena CSS helps us bring competitive, future-oriented solutions to market faster, meeting the demands of software-defined vehicles and the expectations of leading OEMs and tier-one suppliers.”
Editor: Chip Intelligence – Langke Sword
Previous Exciting ArticlesGigaDevice Innovation International Headquarters Settles in SingaporeTSMC’s Wei Zhejia: AI Needs More Capacity, Not Afraid of Technology Leakage in Overseas Factories!Abandon Cortex! Arm Launches 5 New Brands and Plans to Compete with Customers by Developing Its Own Chips!Wolfspeed Plans to File for Bankruptcy, Renesas Dissolves Silicon Carbide Team!Synopsys CEO Sends Internal Letter: New US Regulations Will Affect All Chinese Customers!Zhongwei Company’s Yin Zhiyao: Will Continue to Focus on High-End, Fill Weak Links, Not Afraid of Competitor Competition!Has the US Requested to Cut Off EDA Supply to China? Synopsys and Cadence Stocks Plummet!Test Results Released: Qualcomm’s 5G Baseband Performance Beats Apple’s C1!Has the US Requested Three Major EDA Companies to Cut Off Supply to China? No Notification Received in China!3 Years of Losses Exceed 800 Million! Silicon Carbide IDM Giant Basic Semiconductor Files for Hong Kong Stock Exchange!Benchmarking RTX4060! Lishan Technology’s 6nm High-Performance GPU Successfully Lights Up: Has Secured 100 Million Yuan in Pre-Orders!Not Custom Arm CSS! What Exactly Did Xiaomi’s Xuanjie O1 Develop?Huawei and MediaTek Patent Lawsuit Jurisdiction Dispute, Supreme Court Makes Final Decision!Xiaomi’s 13.5 Billion Burnt on “Xuanjie” Dual Chips, Is It Hard Enough?Jensen Huang Criticizes US AI Chip Restriction Policies, NVIDIA’s Market Share in China Has Dropped to 50%!Wolfspeed to File for Bankruptcy, Stock Price Plummets Over 70%!
For industry communication and collaboration, please add WeChat: icsmart01
Chip Intelligence Official Communication Group: 221807116