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

2024-11-25 21:49:31 +02:00 · 2024-11-25 21:49:31 +02:00 · 106964e3d2
commit 106964e3d2
parent 80acb7b430
1 changed files with 1 additions and 1 deletions
--- a/ggml/src/ggml-metal/ggml-metal.m
+++ b/ggml/src/ggml-metal/ggml-metal.m
@ -1927,7 +1927,7 @@ 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