Edge AI Computing

Run inference where the data is produced: edge AI boxes, accelerator modules and edge controllers for bandwidth-limited, latency-sensitive or data-residency scenarios.