For high-performance inferencing applications on the edge, ASUS provides a solution with this PCIe Gen3 AI accelerator card based on the Google® Coral Edge TPU processors.
The CRL-G18U-P3DF has 8 Google® Coral Edge TPU M.2 modules, for a total performance of 32 TOPS, at a peak power consumption of 36 W.
The Coral TPUs supports the TensorFlow Lite machine learning framework. Each Coral TPU provides up to 4 TOPS. Thanks to the pipelining feature of the Coral platform you can execute different segments of the same model on different Edge TPUs. This way, you can run more complex models efficiently, and accelerate execution even further.
Or simply use this device in the desktop computer of your choice to run ML inferencing on several separate video streams.
The twin turbofans optimize the heat dissipation for the Coral TPUs, allowing your machine learning models to run smoothly without running into thermal issues due to overheating.
Applications include Manufacturing (defect detection, utilities monitoring, safety), Retail (in-store automated checkout, crowd-density analysis, intelligent signage), Transportation (traffic management, fleet management, parking) and Surveillance (intrusion, virtual fencing, security).
Ubuntu 18.04, Debian 10 and Windows are supported as operating systems.
Note: This PCIe expansion card requires a PCI Express 3.0 x 16 Expansion slot. The physical dimensions of the card (42.1 mm height) take up two PCIe slots.
Machine Learning Accelerator:
Interface:
Dimensions & Weight:
Thermals & Operating Temperature:
Additional Legal Information
The products offered by Google are unrelated to the products offered under the CORAL trademarks owned by Orient Development Enterprises Ltd. in Taiwan.