Coral USB Accelerator

The Coral USB Accelerator is a USB accessory that brings machine learning inferencing to existing systems. It works with the Raspberry Pi and Linux, Mac, and Windows systems.

The Accelerator adds an Edge TPU coprocessor to your system, enabling high-speed machine learning inferencing on a wide range of systems, simply by connecting it to a USB port!

Performs High-speed Machine Learning Inferencing

The on-board Edge TPU coprocessor is capable of performing 4 trillion operations (tera-operations) per second (TOPS), using 0.5 watts for each TOPS (2 TOPS per watt). 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. See below section for performance benchmarks.

Supports all major platforms

Connects via USB to any system running Debian Linux (including Raspberry Pi), macOS, or Windows 10.

Supports TensorFlow Lite

No need to build models from the ground up. TensorFlow Lite models can be compiled to run on the Edge TPU.

Supports AutoML Vision Edge

Easily build and deploy fast, high-accuracy custom image classification models to your device with AutoML Vision Edge.

Tech specs

* Compatible with USB 2.0 but inferencing speed is much slower.

Datasheet & Resources

• Datasheet
• Getting started with the USB Accelerator
• USB Accelerator datasheet
• Coral Edge TPU Frequently asked questions (FAQ)
• Edge TPU Python API overview
  • Note: this is NOT the TensorFlow Lite API, but an alternative API intended for users who have not used TensorFlow before and simply want to start with image classification and object detection
• 3D CAD model (.STEP file)
• Edge TPU inferencing overview
• TensorFlow models on the Edge TPU

Performance Benchmarks

An individual Edge TPU is capable of performing 4 trillion operations (tera-operations) per second (TOPS), using 0.5 watts for each TOPS (2 TOPS per watt). How that translates to performance for your application depends on a variety of factors. Every neural network model has different demands, and if you're using the USB Accelerator device, total performance also varies based on the host CPU, USB speed, and other system resources.

With that said, the table below compares the time spent to perform a single inference with several popular models on the Edge TPU. For the sake of comparison, all models running on both CPU and Edge TPU are the TensorFlow Lite versions.

This represents a small selection of model architectures that are compatible with the Edge TPU:

Note: These figures measure the time required to execute the model only. It does not include the time to process input data (such as down-scaling images to fit the input tensor), which can vary between systems and applications. These tests are also performed using C++ benchmark tests, whereas our public Python benchmark scripts may be slower due to overhead from Python.

¹ Desktop CPU: Single 64-bit Intel(R) Xeon(R) Gold 6154 CPU @ 3.00GHz
² Embedded CPU: Quad-core Cortex-A53 @ 1.5GHz
³ Dev Board: Quad-core Cortex-A53 @ 1.5GHz + Edge TPU