AIPU Targets Lower-Power Enterprise AI Inference

Executive Briefing & Key Highlights

What happens when AI inference has to run continuously inside enterprise servers? An accelerator architecture targets higher inference efficiency without replacing existing infrastructure. Axelera AI has launched its Europa AI Processing Unit (AIPU), designed to handle inference workloads in enterprise and data centre environments. The accelerator supports workloads including agentic AI, vision-la