Gluon AI Co-Processor

Gluon, is an 8 TOPS inference co-processor taped-out with 16 nm TSMC FF technology. Gluon is optimized for vision applications and provides the best performance for image classification, face recognition, object detection and other video analytics applications. Alpha ICs provides best-in-class solutions for video analytics in terms of performance, power, and cost. The Gluon PCIe card and evaluation board are available for customer evaluation.
High performance
Gluon AI accelerator is optimized for vision applications and provides maximum throughput with minimum latency and low power.
Ease-of-use
Gluon comes with a complete software tool-chain that enables easy deployment of any trained model on Gluon. One-click quantizer to convert FP32 to INT8 weights.
Versatile
The AI processing is not constrained by neural network size or image resolution. We support different AI frameworks.
Robust
Gluon supports different type of Neural Networks and AI work loads.