PICTURE
Coral USB Accelerator adds an edge of TPU coprocessor to your system, allowing machine learning to come to a wide system, just by connecting it to a USB port.
Makes high-speed ml inferencing
The on-board edge of TPU Coprocessor has the ability to make 4 trillion operations (Terility operations) per second (tops for each upper (2 tops of each State-of-the-art models like MobileNet V2 at around 400 fps, in an effective way. See more polelesscance benchmarks.
Supports all major platforms
Connects through 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. Tensorflow Lite Models can be grown to run along TPU.
Specs in Tech
|
Ml accelerator |
Google Spect TPU Coprocessor: 4 tops (int8); 2 ends per watt |
|
connectors |
USB 3.0 Type-C * (Data / Power) |
|
Dimensions |
65 mm x 30 mm |
| availability |
Australia, European Union (except France, Czech Republic), Ghana, Hong Kong, Indona, Japan, Thailand, Thailand, Thailand, Vietnam, Vietnam, Vietnam, Vietnam, Vietnam, Vietnam, Vietnam, Vietnam, Vietnam, Vietnam |
* Adjustable to USB 2.0 but inferencing speed is slow.
https://lh3.googleusercontent.com/y5YE-KW6BGFdYMJHS8UUktFrD6cow0mbU4pBCQnmO6_5fiBdUHEAGTXwH8QJ-6zfA0iP53C45CC_ur3t33QB-DDubfqomtj9kSFz7g=s0
2025-02-18 17:16:00
