Ara240 DNPU: Edge AI, Accelerated

About This Video

Run advanced LLMs and vision AI natively with the Ara240 DNPU. Delivering up to 40 eTOPS of power-efficient performance, enabling low-latency multimodal AI directly on-device—no cloud required.

Enable next generation AI with the Ara240 DNPU.

Highlights

• Up to 40 eTOPS of AI acceleration for real-time edge inference
• Enables on-device generative AI, LLMs and vision AI without cloud dependency
• Scalable, low-latency performance with high memory bandwidth for advanced AI workloads

Resources

• Ara240 DNPU
• FRDM i.MX 8M Plus Development Board
• FRDM i.MX 95 Development Board
• Ara Software Development Kit
• Ara240 16 GB M.2 Module