AI accelerator (NPU) IP - 32 to 128 TOPS

The Origin™ E8 is designed for applications where performance is a primary design goal, including automotive/ADAS and data center applications. Expedera’s advanced memory management ensures sustained DRAM bandwidth and optimal total system performance. Featuring 32 to 128 TOPS performance with up to 90% real-world utilization (measured on-chip running common workloads such as ResNet), the Origin E8 Deep Learning Accelerator (DLA) excels at image-related tasks like computer vision, image classification, and object detection.

Expedera's scalable tile-based design includes a single controller (SSP), and multiple matrix-math units (MMP), accumulators (PSM), vector engines (VSP) and memory to store the network. Specific configurations depend on unique application requirements. The unified compute pipeline architecture enables highly efficient hardware scheduling and advanced memory management to achieve unsurpassed end-to-end low-latency performance. The patented architecture is mathematically proven to utilize the least amount of memory for neural network (NN) execution. This minimizes die area, reduces DRAM access, improves bandwidth, saves power, and maximizes performance.