metal : enable mat-vec kernels for bs <= 4 (llama/10491)

This commit is contained in:
Georgi Gerganov 2024-11-25 21:49:31 +02:00
parent ac33379a35
commit f6d518fc4c

View File

@ -1935,7 +1935,7 @@ static void ggml_metal_encode_node(
// find the break-even point where the matrix-matrix kernel becomes more efficient compared
// to the matrix-vector kernel
int ne11_mm_min = 1;
int ne11_mm_min = 4;
#if 0
// the numbers below are measured on M2 Ultra for 7B and 13B models