3 points | by JPLeRouzic an hour ago
1 comments
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.
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.