From 425c6bbb6c3236cf7b3757615cb32e2db897e098 Mon Sep 17 00:00:00 2001 From: Iwan Kawrakow Date: Tue, 13 Feb 2024 14:37:16 +0200 Subject: [PATCH] iq1_s: Metal works, but quite slow As usual, Apple Silicon does not like the code I write. --- ggml-metal.metal | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ggml-metal.metal b/ggml-metal.metal index 16b4cc478..235a3c7cf 100644 --- a/ggml-metal.metal +++ b/ggml-metal.metal @@ -4399,7 +4399,7 @@ void kernel_mul_mv_iq1_s_f32_impl( for (int row = 0; row < N_DST; ++row) { all_sum = simd_sum(sumf[row]); if (tiisg == 0) { - dst[r1*ne0 + im*ne0*ne1 + first_row + row] = all_sum * 0.5f; + dst[r1*ne0 + im*ne0*ne1 + first_row + row] = all_sum; } } }