The lineup includes two entirely new products—the X160 Gen 2 and X180 Gen 2—alongside upgraded X280 Gen 2 and X390 Gen 2 and XM Gen 2. The new products include enhanced vector processing capabilities that, unlike traditional processors, are able to handle increasing AI workloads.
The processors target the rapidly growing edge AI market, where companies need efficient processing power for applications such as autonomous robotics, industrial automation and smart IoT devices. They are ideal for audio enhancement, health and fitness monitoring, home security and object detection, and predictive maintenance in industrial systems.
X160 Gen 2 targets edge compute and IoT applications, delivering high efficiency in a compact footprint. X180 Gen 2 provides higher performance for edge inference, AI training, and data center integration.
Traditional systems with accelerators force developers to rely on complex software to coordinate data between AI accelerators and other system components, creating performance delays and increasing power consumption, argues SiFive, whereas SiFive’s new vector-enabled cores speed up AI workloads by serving as dedicated Accelerator Control Units that directly manage AI hardware, streamline data flow and reduce the software overhead that slows down AI applications.
Vector engines process multiple data items in parallel, reducing instruction overhead and power consumption. For edge AI, vector pipelines are especially effective when working with smaller datatypes.
Compared to traditional scalar CPUs, vector-enabled chips can run AI models faster, with a smaller footprint and lower power consumption—ideal for far-edge applications.
The Second Generation deliver four key improvements over previous generations:
New co-processor interface for better AI accelerator integration
Additional new instruction set extensions optimized for AI algorithms.
A novel new design to reduce memory latency
Highly efficient memory subsystem.
From narrow to wide vector registers, to XM with a highly scalable Matrix engine. SiFive’s Second Generation of Intelligence IP offers customers flexibility and many performance, area and power options.
All of the X-Series IPs can be used as a Accelerator Control Unit (ACU), providing accelerator control and assist functions to a customer’s accelerator engine via the specialized co-processor interfaces (SSCI and VCIX).
This allows customers to focus on the data processing capabilities and innovations at the platform level and simplifies the software stack.All five Gen 2 products are available for licensing immediately, with first silicon expected in Q2 2026.
