edgeAI picks Efficient's Electron E1 for next-generation physical AI products

US processor company Efficient Computer announced on 3 November that AI semiconductor company edgeAI has selected its Electron E1 processor to develop next-generation physical AI and critical infrastructure products. The company said the Electron E1 is now in volume production.

Low-power, on-device computing

Efficient says the Electron E1 is designed to run complex AI models on-device, targeting robotics and drones, critical infrastructure observability and industrial wearables, where power, cost and size matter. The company claims the chip offers higher performance and greater energy efficiency than many AI accelerators, and up to 100 times better energy efficiency than low-power CPUs. Those figures are company claims and have not been independently verified.

Existing code carries over

Efficient says its effcc compiler maps customers' existing code to the E1's dataflow architecture without rewriting it, and the processor's programmable architecture allows applications to be updated in software, including over the air, as AI models evolve. edgeAI CEO Yang Jung Woong said the company evaluated a wide range of processors and found none that matched the Electron E1's combination of performance and efficiency, running its models in real time at a fraction of the power.

Company background

Efficient CEO and co-founder Brandon Lucia said the company is scaling production of the Electron E1 as new funding and demand from customers including edgeAI grow; the company recently raised a $97 million Series B round to push its energy-efficiency advantage up the compute stack. Stock image: a computer chip.