M.2 Accelerator A+E key Google Edge TPU

Boost your AI projects with the M.2 Google Coral gas pedal!

The M.2 Google Coral gas pedal is a powerful solution for developers looking to integrate advanced machine learning capabilities into embedded devices. Featuring the Google Edge TPU processor, this gas pedal is designed to run artificial intelligence tasks directly at the edge, without the need for a cloud connection.

Peak performance for your AI projects

With peak performance of 4 TOPS (tera operations per second) for int8 operations, this gas pedal lets you run complex machine learning models with remarkable energy efficiency of 2 TOPS per watt. This power makes the Google Coral ideal for real-time applications such as computer vision, speech recognition and intelligent IoT systems.

Compact, versatile design

The gas pedal is available in M.2 format (with A+E key) with dimensions: 22 x 30 mm. This format makes it easy to integrate the gas pedal into a variety of embedded systems and space-saving devices. It uses a PCIe Gen2 x1 interface, guaranteeing compatibility with a wide range of peripherals, including the Pineberry card.

Integrated power management and thermal robustness

Thanks to integrated power management, this gas pedal optimizes its energy consumption, enabling you to maximize the autonomy of your devices. What's more, it can operate over a wide temperature range, from -20 to +85°C, making it suitable for a wide variety of environments, even the most demanding.

Conclusion

The Google Coral M.2 is an ideal solution for those looking to deploy high-performance AI models directly at the edge. Compact, energy-efficient and robust, it is a valuable asset for the development of intelligent embedded systems.

Useful documentation

Benchmarks