Model Compression based on Quantization, Pruning, Knowledge Distillation, and AutoML

Opt-AI’s optimization solution compresses AI models using state-of-the-art techniques
including quantization, pruning, knowledge distillation, and AutoML.



Knowledge Distillation

Moreover, AI models are accelerated on any target AI chipset including CPUs, GPUs, ASICs and FPGAs,
with the Opt-AI’s hardware-aware automated profiling solution

Inference Speedup

Accuracy Preserving

Hardware Full-Awareness

Model size & Memory Reduction