The Coral M.2 Accelerator featuring Dual Edge TPU comprises an M.2 module (E-key) equipped with two Edge TPU ML accelerators, each possessing its own PCIe Gen2 x1 interface. Developed by Google, the Edge TPU is a compact ASIC engineered to expedite TensorFlow Lite models with exceptional energy efficiency, boasting a capacity of 4 trillion operations per second (4 TOPS) while consuming a mere 2 watts of power—equating to 2 TOPS per watt. For instance, a single Edge TPU can handle advanced mobile vision models like MobileNet v2 at nearly 400 frames per second. This on-device ML processing not only diminishes latency and enhances data privacy but also eliminates the necessity for a constant internet connection. With the inclusion of two Edge TPUs within this module, the potential for inferences per second doubles (8 TOPS), offering various avenues for optimization such as executing two models concurrently or implementing a model pipeline across both Edge TPUs.
Key Features:
Performs high-speed ML inferencing
Each Edge TPU coprocessor is capable of performing 4 trillion operations per second (4 TOPS), using 2 watts of power. For example, it can execute state-of-the-art mobile vision models such as MobileNet v2 at almost 400 FPS, in a power efficient manner.
With the two Edge TPUs in this module, you can double the inferences per second (8 TOPS) in several ways, such as by running two models in parallel or pipelining one model across both Edge TPUs.
Works with Debian Linux and Windows
Integrates with Debian-based Linux or Windows 10 systems with a compatible card module slot.
No need to build models from the ground up. TensorFlow Lite models can be compiled to run on the Edge TPU.
Warranty Period: 12 months
International
Singapore
Malaysia
Thailand
Vietnam
