SiFive introduces the new 2nd Generation Intelligence family, featuring advanced processors tailored for AI workloads and edge computing applications in various sectors such as automotive, robotics, industrial infrastructure, and IoT infrastructure.
These new products enhance scalar, vector, and matrix processing performance to meet the demands of modern AI-driven technologies.
The X160 Gen 2 and X180 Gen 2 models from SiFive’s latest lineup are specifically crafted for far-edge and IoT scenarios, offering high throughput capabilities in compact, power-constrained environments.
Patrick Little, the CEO of SiFive, emphasizes the pivotal role of AI in driving the evolution of RISC-V technology. He highlights the growing interest in the X100 series among prominent semiconductor companies and underscores the enhanced features and configurability of the new 2nd Generation Intelligence IP for accelerating customer designs.
SiFive’s RISC-V vector-based IPs are engineered to optimize AI model performance while minimizing power consumption and space requirements, making them ideal for edge AI applications.
The X-Series IP family, including Accelerator Control Units (ACU) functionality, streamlines data processing tasks and reduces software complexities, ultimately enabling differentiated platform-level performance.
Addressing the surging demand for AI capabilities, SiFive facilitates high-performance computing solutions, particularly in the realm of AI edge computing, which is projected to witness significant growth.
Licensing options for the new products are now available, with initial silicon expected to be available in the second quarter of 2026, marking a significant advancement in unifying compute architectures for AI at the edge.
This product launch signifies a critical stride towards delivering scalable, energy-efficient AI computing solutions that span from microcontrollers at the far edge to data center accelerators, ushering in a new era of cohesive AI architecture.
Related
AI hardware | data center | edge AI | edge computing | RISC-V | semiconductor | SiFive
