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

diff --git a/ggml/src/ggml-metal/ggml-metal.m b/ggml/src/ggml-metal/ggml-metal.m
index 63baaf163581c69e00e58e43c427950f2404789b..c47f07a9e041681e53b6ddd27d6db505ae02dc11 100644 (file)
--- 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