Developed a custom GEMM (General Matrix Multiply) IP core using, implementing Int8 Quantization with on-chip de-quantization to reduce memory. - View it on GitHub