Google continues to expand its range of AI products and services with a trio of new hardware devices aimed at the development community.
The devices don’t seem to have been officially announced yet and were first spotted by Hackster. They’re being introduced under a new Google Coral brand (which is itself still “in beta”), and include a development board that sells for $149.99, a USB accelerator that goes for $74.99, and a 5-megapixel camera that’s available for $24.99.
Both dev board and accelerator are powered by Google’s Edge TPU chips, which are ASIC processors no bigger than your fingernail that are designed to run AI models without breaking a sweat. The camera, meanwhile, is as an add-on for the dev board.
None of these products are good for training machine learning algorithms. That’s still a heavy-duty task that requires GPUs or cloud processing power. Instead, they facilitate the “inference” part of AI, which comes after you’ve trained your model and are using it in the wild. To that end, both the accelerator and dev board support TensorFlow Lite, a lightweight version of the company’s AI framework that’s designed for mobile and IoT devices.
You can, of course, run AI models on any old CPU, but they’ll be slow and cranky. Specially designed chips like Google’s TPU processors do the job much more efficiently without the need for a dedicated internet connection or chunky power source.
The USB accelerator can boost inference on any Linux machine, while the dev board’s array of pins and ports make it perfect for prototyping hardware and other experimental applications. (Hackaday notes that its setup is almost identical to that of the Raspberry Pi.)
Google isn’t the first company to offer this sort of product to developers, as Intel released its own USB accelerator a few years ago, the Movidius-powered Neural Compute Stick. Products like this help companies reach out to the developer community, boosting adoption of their other AI services.
For more information on these devices, you can find full spec sheets and resources, including tutorials and test models, over on Google’s new Coral homepage.