diff --git a/CMakeLists.txt b/CMakeLists.txt
index b871b3236..178d693a6 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -75,7 +75,7 @@ endif()
 
 file(GLOB GGML_SOURCES_CUDA "ggml/src/ggml-cuda/*.cu")
 list(APPEND GGML_SOURCES_CUDA "ggml/src/ggml-cuda/ggml-cuda.cu")
-file(GLOB SRCS "ggml/src/ggml-cuda/template-instances/fattn-wmma*.cu")
+file(GLOB SRCS "ggml/src/ggml-cuda/template-instances/fattn-mma*.cu")
 list(APPEND GGML_SOURCES_CUDA ${SRCS})
 file(GLOB SRCS "ggml/src/ggml-cuda/template-instances/mmq*.cu")
 list(APPEND GGML_SOURCES_CUDA ${SRCS})
diff --git a/Makefile b/Makefile
index 970d86b01..af898885d 100644
--- a/Makefile
+++ b/Makefile
@@ -167,7 +167,7 @@ endif
 endif
 
 # it is recommended to use the CMAKE file to build for cublas if you can - will likely work better
-OBJS_CUDA_TEMP_INST = $(patsubst %.cu,%.o,$(wildcard ggml/src/ggml-cuda/template-instances/fattn-wmma*.cu))
+OBJS_CUDA_TEMP_INST = $(patsubst %.cu,%.o,$(wildcard ggml/src/ggml-cuda/template-instances/fattn-mma*.cu))
 OBJS_CUDA_TEMP_INST += $(patsubst %.cu,%.o,$(wildcard ggml/src/ggml-cuda/template-instances/mmq*.cu))
 OBJS_CUDA_TEMP_INST += $(patsubst %.cu,%.o,$(wildcard ggml/src/ggml-cuda/template-instances/fattn-vec*q4_0-q4_0.cu))
 OBJS_CUDA_TEMP_INST += $(patsubst %.cu,%.o,$(wildcard ggml/src/ggml-cuda/template-instances/fattn-vec*q8_0-q8_0.cu))