diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index c6aecec6e94d59..e569981989adec 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -660,10 +660,26 @@ jobs:
         run: |
           brew update
 
+      - name: Build llama.cpp with CMake
+        id: cmake_build
+        run: |
+          sysctl -a
+          mkdir build
+          cd build
+          cmake -G Xcode .. \
+            -DGGML_METAL_USE_BF16=ON \
+            -DGGML_METAL_EMBED_LIBRARY=ON \
+            -DLLAMA_BUILD_EXAMPLES=OFF \
+            -DLLAMA_BUILD_TESTS=OFF \
+            -DLLAMA_BUILD_SERVER=OFF \
+            -DCMAKE_OSX_ARCHITECTURES="arm64;x86_64"
+          cmake --build . --config Release -j $(sysctl -n hw.logicalcpu)
+          cmake --install . --config Release
+
       - name: xcodebuild for swift package
         id: xcodebuild
         run: |
-          xcodebuild -scheme llama -destination "${{ matrix.destination }}"
+          xcodebuild -scheme llama-Package -destination "${{ matrix.destination }}"
 
       - name: Build Swift Example
         id: make_build_swift_example
diff --git a/Package.swift b/Package.swift
index d9e8a4e2d21d7b..01c996d2420376 100644
--- a/Package.swift
+++ b/Package.swift
@@ -2,57 +2,6 @@
 
 import PackageDescription
 
-var sources = [
-    "src/llama.cpp",
-    "src/llama-vocab.cpp",
-    "src/llama-grammar.cpp",
-    "src/llama-sampling.cpp",
-    "src/unicode.cpp",
-    "src/unicode-data.cpp",
-    "ggml/src/ggml.c",
-    "ggml/src/ggml-aarch64.c",
-    "ggml/src/ggml-alloc.c",
-    "ggml/src/ggml-backend.cpp",
-    "ggml/src/ggml-backend-reg.cpp",
-    "ggml/src/ggml-cpu/ggml-cpu.c",
-    "ggml/src/ggml-cpu/ggml-cpu.cpp",
-    "ggml/src/ggml-cpu/ggml-cpu-aarch64.c",
-    "ggml/src/ggml-cpu/ggml-cpu-quants.c",
-    "ggml/src/ggml-threading.cpp",
-    "ggml/src/ggml-quants.c",
-]
-
-var resources: [Resource] = []
-var linkerSettings: [LinkerSetting] = []
-var cSettings: [CSetting] =  [
-    .unsafeFlags(["-Wno-shorten-64-to-32", "-O3", "-DNDEBUG"]),
-    .unsafeFlags(["-fno-objc-arc"]),
-    .headerSearchPath("ggml/src"),
-    // NOTE: NEW_LAPACK will required iOS version 16.4+
-    // We should consider add this in the future when we drop support for iOS 14
-    // (ref: ref: https://developer.apple.com/documentation/accelerate/1513264-cblas_sgemm?language=objc)
-    // .define("ACCELERATE_NEW_LAPACK"),
-    // .define("ACCELERATE_LAPACK_ILP64")
-]
-
-#if canImport(Darwin)
-sources.append("ggml/src/ggml-common.h")
-sources.append("ggml/src/ggml-metal/ggml-metal.m")
-resources.append(.process("ggml/src/ggml-metal/ggml-metal.metal"))
-linkerSettings.append(.linkedFramework("Accelerate"))
-cSettings.append(
-    contentsOf: [
-        .define("GGML_USE_ACCELERATE"),
-        .define("GGML_USE_METAL"),
-        .define("GGML_USE_CPU")
-    ]
-)
-#endif
-
-#if os(Linux)
-    cSettings.append(.define("_GNU_SOURCE"))
-#endif
-
 let package = Package(
     name: "llama",
     platforms: [
@@ -65,26 +14,6 @@ let package = Package(
         .library(name: "llama", targets: ["llama"]),
     ],
     targets: [
-        .target(
-            name: "llama",
-            path: ".",
-            exclude: [
-               "build",
-               "cmake",
-               "examples",
-               "scripts",
-               "models",
-               "tests",
-               "CMakeLists.txt",
-               "Makefile",
-               "ggml/src/ggml-metal-embed.metal"
-            ],
-            sources: sources,
-            resources: resources,
-            publicHeadersPath: "spm-headers",
-            cSettings: cSettings,
-            linkerSettings: linkerSettings
-        )
-    ],
-    cxxLanguageStandard: .cxx11
+        .systemLibrary(name: "llama", pkgConfig: "llama"),
+    ]
 )
diff --git a/Sources/llama/llama.h b/Sources/llama/llama.h
new file mode 100644
index 00000000000000..41725880ed8c06
--- /dev/null
+++ b/Sources/llama/llama.h
@@ -0,0 +1,4 @@
+#pragma once
+
+#include <llama.h>
+
diff --git a/Sources/llama/module.modulemap b/Sources/llama/module.modulemap
new file mode 100644
index 00000000000000..d010555b1cb658
--- /dev/null
+++ b/Sources/llama/module.modulemap
@@ -0,0 +1,5 @@
+module llama [system] {
+    header "llama.h"
+    link "llama"
+    export *
+}
diff --git a/cmake/llama.pc.in b/cmake/llama.pc.in
index 326acbb6108fd0..0b2b6bcfabfd1b 100644
--- a/cmake/llama.pc.in
+++ b/cmake/llama.pc.in
@@ -6,5 +6,5 @@ includedir=${prefix}/include
 Name: llama
 Description: Port of Facebook's LLaMA model in C/C++
 Version: @PROJECT_VERSION@
-Libs: -L${libdir} -lllama
+Libs: -L${libdir} -lggml  -lggml-base -lllama
 Cflags: -I${includedir}