"The recent availability of small and low power super computers makes edge intelligence attractive for a lot of applications. However, edge intelligence requires specific hardware and this makes the rollout of new edge AI applications challenging. Therefore AI-BLOX has developed modular hardware blox to accelerate your edge AI application rollout."
The modular take on hardware that AI-BLOX brings to the table nicely aligns with our modular approach to edge AI software: AI-BLOX separates accelerators and physical I/O modularly from the surrounding platform making it easy to quickly iterate edge AI solutions. The Scailable AI Manager separates the AI inference pipeline both from the underlying hardware (no need to compile your models specifically for your target), and from the AI model itself. This allows for extremely fast iterations of AI models on the edge by providing a clear separation of the model development and (embedded) pipeline development cycles.
You can easily install the AI manager on AI-Blox devices using our simple one-line install. Connect to the device and open a shell. Next, type the following command:
sudo bash -ic "$(wget -q -O - https://get.sclbl.net)"
This will install the Scailable AI manager on the device.
If any NVIDIA accelerators are found during the installation process you will be prompted to add support for the accelaration within the pipeline.
AI-BLOX are modular; depending on your use-case you can tailor the device to fit your hardware resource needs. With the Scailable platform it is easy to profile your full AI pipeline on the device and pick the hardware your need.