iEi Mustang-M2BM-MX2 acceleration card

Click2Contact

iEi’s Mustang-M2BM-MX2 is a deep learning convolutional neural network acceleration card for speeding up AI inference in a flexible and scalable way. Equipped with two Intel Movidius Myriad X vision processing units (VPUs), the card can help enable high-performance computing in an existing system.

The VPUs can help run AI faster and are suitable for low power consumption applications such as surveillance, retail and transportation. With the advantage of power efficiency and high performance to dedicate DNN topologies, the product can be implemented in AI edge computing devices to reduce total power usage, providing long duty time for rechargeable edge computing equipment.

The product is powered by Intel’s OpenVINO toolkit, which is based on convolutional neural networks (CNN) and extends workloads across Intel hardware, maximising performance. It can optimise pre-trained deep learning models such as Caffe, MXNET and Tensorflow into an IR binary file then execute the inference engine across Intel hardware heterogeneously, such as CPU, GPU, Intel Movidius Myriad X VPU and FPGA.

Other features include: an M.2 BM key form factor (22 x 80 mm); power efficiency of ~7.5 W; and an operating temperature of -20 to +60°C.