TechWatch

Arm China points Zhouyi X3 at the edge

Arm China launches Zhouyi X3 NPU IP for large models at the edge.

David Harold
Arm China

Arm China has introduced its Zhouyi X3 NPU IP, a DSP+DSA-based accelerator aimed at bringing large model inference to infrastructure, automotive, mobile, and IoT SoCs. The IP delivers 8–80 FP8 TFLOPS per cluster, with significantly higher on-core bandwidth and improved utilization on LLM workloads. A companion Compass software stack targets one-click deployment from frameworks like PyTorch and Hugging Face, plus extensive quantization support for cloud-to-edge migration. The launch is also the first big expression of Arm China’s “All in AI” strategy and a clear bid to anchor domestic GenAI designs on local IP. (Source: Arm China) While keenly pointing out
...

Enjoy full access with a TechWatch subscription!

TechWatch is the front line of JPR information gathering service, comprising current stories of interest to the graphics industry spanning the core areas of graphics hardware and software, workstations, gaming, and design.

A subscription to TechWatch includes 4 hours of consulting time to be used over the course of the subscription.

Already a subscriber? Login below

This content is restricted

Subscribe to TechWatch