From 3fa98d3213cc57c3558b610a9b1e134af2938ebd Mon Sep 17 00:00:00 2001 From: Kawrakow Date: Sun, 29 Mar 2026 14:36:09 +0200 Subject: [PATCH] Enable all CPU-backend FA supported quants by default (#1549) --- ggml/CMakeLists.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ggml/CMakeLists.txt b/ggml/CMakeLists.txt index c8987b2f..29ec48de 100644 --- a/ggml/CMakeLists.txt +++ b/ggml/CMakeLists.txt @@ -139,7 +139,7 @@ set (GGML_CUDA_COMPRESSION_MODE "size" CACHE STRING set_property(CACHE GGML_CUDA_COMPRESSION_MODE PROPERTY STRINGS "none;speed;balance;size") option(GGML_IQK_FLASH_ATTENTION "ggml: enable the IQK FlashAttention CPU kernels" ON) -option(GGML_IQK_FA_ALL_QUANTS "ggml: compile all quants for IQK FlashAttention" OFF) +option(GGML_IQK_FA_ALL_QUANTS "ggml: compile all quants for IQK FlashAttention" ON) option(GGML_CURL "ggml: use libcurl to download model from an URL" OFF) option(GGML_HIPBLAS "ggml: use hipBLAS" OFF)