no code implementations • 15 Feb 2024 • Kyle Marino, Pengmiao Zhang, Viktor Prasanna
We evaluate ME-ViT on systolic array sizes of 32 and 16, achieving up to a 9. 22$\times$ and 17. 89$\times$ overall improvement in memory bandwidth, and a 2. 16$\times$ improvement in throughput per DSP for both designs over state-of-the-art ViT accelerators on FPGA.