SiFive announced a strategic partnership with Edge AI innovator Kinara to launch the HiFive Xara X280, a compact USB-based enablement board that provides developers with direct, bare-metal access to the powerful SiFive Intelligence X280 processor. This new platform is designed to accelerate the development of RISC-V vector software and edge AI applications.
At the heart of the Xara board is Kinara’s Ara-2 processor, which integrates two 64-bit SiFive X280 RISC-V vector cores. These cores are optimized for high-performance compute tasks such as tensor filtering, floating-point operations, and real-time data processing. The X280 cores are tightly coupled with Kinara’s Neural Processing Unit cores, delivering up to 40 TOPS of AI inference performance. This architecture supports a wide range of AI workloads, including convolutional neural networks, generative AI models, and multimodal vision transformers like LLaVA and LLaMA.
“SiFive’s RISC-V cores are a perfect complement to Kinara’s neural processor cores, enabling performance- and cost-optimized end-to-end inference for edge AI use cases,” said Ravi Annavajjhala, CEO of Kinara. “Together with SiFive, the combined solution is capable of supporting a range of AI workloads, from traditional CNNs to advanced Generative AI and multi-modal vision transformers.”
The HiFive Xara X280 board is designed to give developers early access to the X280 IP, enabling them to evaluate its real-time behavior and performance before integrating it into custom silicon designs. The board also includes sample code and development tools to streamline the software development process.
“We designed the Xara board with the intention of allowing SiFive customers to evaluate the real-time behavior of the X280 IP before integrating it into their own custom chips,” said Jack Kang, SVP of Worldwide Business Development, Sales, and Customer Experience at SiFive. “The expanded ability to access Kinara's Ara-2 processor means developers can also explore the latest in edge AI processing capabilities.”
