TENSAI® Neural Network Compiler

The TENSAI Flow exclusive neural network compiler delivers the best optimization for neural networks running on Eta Compute’s SoC as well as the industry’s best power efficiency.

TENSAI neural network compiler addresses the two critical aspects of device-level optimization for edge devices: memory and efficiency.

Memory optimization

  • Operation pipelining, redundancy reduction
  • Dynamic allocation and deallocation

Speed and efficiency optimization

  • Optimized kernels for Cortex-M and DSP accelerator
  • Multicore workload balancing

Best Efficiency for AI at the edge

Compared to direct implementation on a leading competitor device of the same CIFAR10 neural network, the TENSAI neural network compiler on TENSAI SoC improves energy per inference by a factor 54x. Using the CIFAR10 neural network from TENSAI neural network zoo and TENSAI neural network compiler improves the energy per inference further, bringing it to a staggering 200x factor.