Comment Comment by JPLeRouzic The author says:"It used Voronoi Cells and reduced most math to 8- and 16-bit integer calculations with one or two single-precision floating-point calculations."I wonder if it can be used to speed up LLM inference. parent JPLeRouzic 1 month ago
"It used Voronoi Cells and reduced most math to 8- and 16-bit integer calculations with one or two single-precision floating-point calculations."
I wonder if it can be used to speed up LLM inference.