Grid Games: The Power of Multiple Grids for Quantizing Large Language Models

ArXi:2605.12327v1 Announce Type: new A major recent advance in quantization is given by microscaled 4-bit formats such as NVFP4 and MXFP4, quantizing values into small groups sharing a scale, assuming a fixed floating-point grid. In this paper, we study the following natural extension: assume that, for each group of values, we are free to select the "better" among two or 4-bit grids marked by one or bits in the scale value.