The world’s first sparsity-native AI System-on-Chip (SoC) GrAI VIP is optimized for ultra-low latency and low power processing at the endpoint to enable Life-Ready AI.
GrAI VIP drastically reduces application latency, for instance, it can reduce the end-to-end latencies for deep learning networks such as Resnet-50 to the order of a millisecond. The GrAI VIP edge AI processor is based on GML’s innovative NeuronFlow™ technology that combines the dynamic Dataflow paradigm with sparse computing to produce massively parallel in-network processing.