diff --git a/CMakeLists.txt b/CMakeLists.txt
index d6c1b3b33d6a2..3ec753b3579d4 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -36,6 +36,12 @@ endif()
 # Option list
 #
 
+if (APPLE)
+    set(LLAMA_METAL_DEFAULT ON)
+else()
+    set(LLAMA_METAL_DEFAULT OFF)
+endif()
+
 # general
 option(LLAMA_STATIC                     "llama: static link libraries"                          OFF)
 option(LLAMA_NATIVE                     "llama: enable -march=native flag"                      OFF)
@@ -76,7 +82,7 @@ option(LLAMA_CUDA_F16                        "llama: use 16 bit floats for some
 set(LLAMA_CUDA_KQUANTS_ITER "2" CACHE STRING "llama: iters./thread per block for Q2_K/Q6_K")
 option(LLAMA_HIPBLAS                         "llama: use hipBLAS"                               OFF)
 option(LLAMA_CLBLAST                         "llama: use CLBlast"                               OFF)
-option(LLAMA_METAL                           "llama: use Metal"                                 OFF)
+option(LLAMA_METAL                           "llama: use Metal"                                 ${LLAMA_METAL_DEFAULT})
 option(LLAMA_MPI                             "llama: use MPI"                                   OFF)
 option(LLAMA_K_QUANTS                        "llama: use k-quants"                              ON)
 option(LLAMA_QKK_64                          "llama: use super-block size of 64 for k-quants"   OFF)
@@ -158,6 +164,31 @@ if (APPLE AND LLAMA_ACCELERATE)
     endif()
 endif()
 
+if (LLAMA_METAL)
+    find_library(FOUNDATION_LIBRARY         Foundation              REQUIRED)
+    find_library(METAL_FRAMEWORK            Metal                   REQUIRED)
+    find_library(METALKIT_FRAMEWORK         MetalKit                REQUIRED)
+
+    message(STATUS "Metal framework found")
+
+    set(GGML_SOURCES_METAL ggml-metal.m ggml-metal.h)
+
+    add_compile_definitions(GGML_USE_METAL)
+    #add_compile_definitions(GGML_METAL_NDEBUG)
+
+    # get full path to the file
+    #add_compile_definitions(GGML_METAL_DIR_KERNELS="${CMAKE_CURRENT_SOURCE_DIR}/")
+
+    # copy ggml-metal.metal to bin directory
+    configure_file(ggml-metal.metal bin/ggml-metal.metal COPYONLY)
+
+    set(LLAMA_EXTRA_LIBS ${LLAMA_EXTRA_LIBS}
+        ${FOUNDATION_LIBRARY}
+        ${METAL_FRAMEWORK}
+        ${METALKIT_FRAMEWORK}
+        )
+endif()
+
 if (LLAMA_BLAS)
     if (LLAMA_STATIC)
         set(BLA_STATIC ON)
@@ -293,29 +324,6 @@ if (LLAMA_CUBLAS)
     endif()
 endif()
 
-if (LLAMA_METAL)
-    find_library(FOUNDATION_LIBRARY         Foundation              REQUIRED)
-    find_library(METAL_FRAMEWORK            Metal                   REQUIRED)
-    find_library(METALKIT_FRAMEWORK         MetalKit                REQUIRED)
-
-    set(GGML_SOURCES_METAL ggml-metal.m ggml-metal.h)
-
-    add_compile_definitions(GGML_USE_METAL)
-    #add_compile_definitions(GGML_METAL_NDEBUG)
-
-    # get full path to the file
-    #add_compile_definitions(GGML_METAL_DIR_KERNELS="${CMAKE_CURRENT_SOURCE_DIR}/")
-
-    # copy ggml-metal.metal to bin directory
-    configure_file(ggml-metal.metal bin/ggml-metal.metal COPYONLY)
-
-    set(LLAMA_EXTRA_LIBS ${LLAMA_EXTRA_LIBS}
-        ${FOUNDATION_LIBRARY}
-        ${METAL_FRAMEWORK}
-        ${METALKIT_FRAMEWORK}
-        )
-endif()
-
 if (LLAMA_MPI)
     cmake_minimum_required(VERSION 3.10)
     find_package(MPI)
diff --git a/Makefile b/Makefile
index b750540fe349e..67afa762c053d 100644
--- a/Makefile
+++ b/Makefile
@@ -1,6 +1,10 @@
 # Define the default target now so that it is always the first target
 BUILD_TARGETS = main quantize quantize-stats perplexity embedding vdot train-text-from-scratch convert-llama2c-to-ggml simple save-load-state server embd-input-test gguf llama-bench baby-llama beam-search tests/test-c.o
 
+ifndef LLAMA_NO_METAL
+BUILD_TARGETS += metal
+endif
+
 # Binaries only useful for tests
 TEST_TARGETS = tests/test-llama-grammar tests/test-grammar-parser tests/test-double-float tests/test-grad0 tests/test-opt tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling tests/test-tokenizer-0-llama tests/test-tokenizer-0-falcon tests/test-tokenizer-1
 
@@ -204,14 +208,24 @@ endif
 endif
 
 ifndef LLAMA_NO_ACCELERATE
-	# Mac M1 - include Accelerate framework.
-	# `-framework Accelerate` works on Mac Intel as well, with negliable performance boost (as of the predict time).
+	# Mac OS - include Accelerate framework.
+	# `-framework Accelerate` works both with Apple Silicon and Mac Intel
 	ifeq ($(UNAME_S),Darwin)
 		CFLAGS  += -DGGML_USE_ACCELERATE
 		LDFLAGS += -framework Accelerate
 	endif
 endif # LLAMA_NO_ACCELERATE
 
+ifndef LLAMA_NO_METAL
+	# By default - use GPU acceleration on Mac OS
+	ifeq ($(UNAME_S),Darwin)
+		CFLAGS   += -DGGML_USE_METAL #-DGGML_METAL_NDEBUG
+		CXXFLAGS += -DGGML_USE_METAL
+		LDFLAGS  += -framework Foundation -framework Metal -framework MetalKit
+		OBJS     += ggml-metal.o
+	endif
+endif # LLAMA_NO_METAL
+
 ifdef LLAMA_MPI
 	CFLAGS += -DGGML_USE_MPI -Wno-cast-qual
 	CXXFLAGS += -DGGML_USE_MPI -Wno-cast-qual
@@ -321,17 +335,10 @@ ggml-cuda.o: ggml-cuda.cu ggml-cuda.h
 	$(HIPCC) $(CXXFLAGS) $(HIPFLAGS) -x hip -c -o $@ $<
 endif # LLAMA_HIPBLAS
 
-ifdef LLAMA_METAL
-	CFLAGS   += -DGGML_USE_METAL #-DGGML_METAL_NDEBUG
-	CXXFLAGS += -DGGML_USE_METAL
-	LDFLAGS  += -framework Foundation -framework Metal -framework MetalKit
-	OBJS     += ggml-metal.o
-endif # LLAMA_METAL
-
-ifdef LLAMA_METAL
+ifndef LLAMA_NO_METAL
 ggml-metal.o: ggml-metal.m ggml-metal.h
 	$(CC) $(CFLAGS) -c $< -o $@
-endif # LLAMA_METAL
+endif # LLAMA_NO_METAL
 
 ifdef LLAMA_MPI
 ggml-mpi.o: ggml-mpi.c ggml-mpi.h
@@ -449,11 +456,7 @@ baby-llama: examples/baby-llama/baby-llama.cpp ggml.o llama.o common.o $(OBJS)
 beam-search: examples/beam-search/beam-search.cpp build-info.h ggml.o llama.o common.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-ifneq '' '$(or $(filter clean,$(MAKECMDGOALS)),$(LLAMA_METAL))'
-BUILD_TARGETS += metal
-endif
-
-ifdef LLAMA_METAL
+ifndef LLAMA_NO_METAL
 metal: examples/metal/metal.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $^ -o $@ $(LDFLAGS)
 endif