Cerence announced a strategic partnership with Arm to enhance the performance of its embedded small language model, CaLLM Edge, by leveraging Arm's Kleidi software library. The collaboration addresses the challenges automakers face in integrating AI capabilities into vehicles, particularly in optimizing CPU performance and maximizing compute power for handling large language models.
With more than 90% of global automakers using Arm technology, the partnership between Cerence and Arm aims to provide a foundational compute architecture that enhances AI capabilities throughout vehicles. Arm's Kleidi software is tailored to accelerate machine learning and optimize neural network operations on Arm-based devices, thereby enabling efficient real-time language processing at the edge.
The partnership facilitates the distribution of AI computation loads between CPUs and graphics processing units (GPUs), which improves the speed and performance of CaLLM Edge. This improvement also supports enhanced privacy and data security for in-car language models, even in the absence of cloud connectivity.
"We are excited to partner with Arm to take CaLLM Edge to the next level, setting new standards for performance and efficiency in edge computing in the car," said Nils Schanz, executive vice president, Product & Technology, Cerence AI. "By combining our expertise in AI-powered language models with Arm’s innovative library, we are continuing our journey to create a new era of voice-first experiences and next-generation AI applications in the automotive space, empowering consumers with smarter, faster, and more responsive in-car assistants."
Suraj Gajendra, vice president of Automotive Products and Software Solutions, Automotive Line of Business at Arm, emphasized the importance of optimizing AI for seamless operations at the edge, noting that Cerence AI is already seeing significant enhancements by utilizing Arm Kleidi. The collaboration is focused on rapidly enabling new, application-specific AI models for vehicles, marking a significant step in the evolution of sophisticated in-vehicle features.
This content may be AI-assisted and is composed, reviewed, edited and approved by S&P Global.
