From e5e72562c5d20b4de175ae6e387b91389e90eaa6 Mon Sep 17 00:00:00 2001 From: Iwan Kawrakow Date: Sun, 3 Mar 2024 18:50:26 +0200 Subject: [PATCH] iq3_s_mult: back to blocks of 32 --- ggml-quants.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ggml-quants.h b/ggml-quants.h index 4ad5d69e7..cb7af5961 100644 --- a/ggml-quants.h +++ b/ggml-quants.h @@ -201,7 +201,7 @@ typedef struct { static_assert(sizeof(block_iq3_xxs) == sizeof(ggml_fp16_t) + 3*(QK_K/8), "wrong iq3_xxs block size/padding"); // 3.4375 bpw -#define IQ3S_BLOCK_SIZE 16 +#define IQ3S_BLOCK_SIZE 32 #if QK_K == 64 #define IQ3S_N_SCALE 2 #else