diff --git a/CMakeLists.txt b/CMakeLists.txt
index caea48c5060..864743bd2a1 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -215,6 +215,15 @@ if (LLAMA_BUILD_COMMON AND LLAMA_BUILD_TOOLS)
     add_subdirectory(tools)
 endif()
 
+# Standalone libmtmd build (e.g. for Apple XCFramework packaging) without pulling
+# in the rest of the tools/ tree. When the full tools build is already enabled,
+# mtmd is built by the tools/ subdirectory above; this hook only fires when
+# LLAMA_BUILD_TOOLS is OFF to avoid double-adding the target.
+option(LLAMA_BUILD_MTMD "llama: build tools/mtmd library standalone" OFF)
+if (LLAMA_BUILD_MTMD AND NOT (LLAMA_BUILD_COMMON AND LLAMA_BUILD_TOOLS))
+    add_subdirectory(tools/mtmd)
+endif()
+
 # Automatically add all files from the 'licenses' directory
 file(GLOB EXTRA_LICENSES "${CMAKE_SOURCE_DIR}/licenses/LICENSE-*")
 
diff --git a/build-xcframework.sh b/build-xcframework.sh
index c25a1ef28c1..295236b90d4 100755
--- a/build-xcframework.sh
+++ b/build-xcframework.sh
@@ -11,6 +11,7 @@ LLAMA_BUILD_EXAMPLES=OFF
 LLAMA_BUILD_TOOLS=OFF
 LLAMA_BUILD_TESTS=OFF
 LLAMA_BUILD_SERVER=OFF
+LLAMA_BUILD_MTMD=ON
 GGML_METAL=ON
 GGML_METAL_EMBED_LIBRARY=ON
 GGML_BLAS_DEFAULT=ON
@@ -35,6 +36,7 @@ COMMON_CMAKE_ARGS=(
     -DLLAMA_BUILD_TOOLS=${LLAMA_BUILD_TOOLS}
     -DLLAMA_BUILD_TESTS=${LLAMA_BUILD_TESTS}
     -DLLAMA_BUILD_SERVER=${LLAMA_BUILD_SERVER}
+    -DLLAMA_BUILD_MTMD=${LLAMA_BUILD_MTMD}
     -DGGML_METAL_EMBED_LIBRARY=${GGML_METAL_EMBED_LIBRARY}
     -DGGML_BLAS_DEFAULT=${GGML_BLAS_DEFAULT}
     -DGGML_METAL=${GGML_METAL}
@@ -123,6 +125,13 @@ setup_framework_structure() {
     cp ggml/include/ggml-blas.h    ${header_path}
     cp ggml/include/gguf.h         ${header_path}
 
+    # Note on mtmd: libmtmd.a is combined into the framework binary (see
+    # combine_static_libraries) so the symbols are exported, but mtmd.h and
+    # mtmd-helper.h are NOT copied into the module because they contain C++
+    # constructs (method-bearing structs, deleters) that break Clang's pure-C
+    # module-map precompilation path. Consumers that need the mtmd ABI should
+    # declare the symbols themselves via an extern C shim header.
+
     # Create module map (common for all platforms)
     cat > ${module_path}module.modulemap << EOF
 framework module llama {
@@ -250,6 +259,7 @@ combine_static_libraries() {
         "${base_dir}/${build_dir}/ggml/src/${release_dir}/libggml-cpu.a"
         "${base_dir}/${build_dir}/ggml/src/ggml-metal/${release_dir}/libggml-metal.a"
         "${base_dir}/${build_dir}/ggml/src/ggml-blas/${release_dir}/libggml-blas.a"
+        "${base_dir}/${build_dir}/tools/mtmd/${release_dir}/libmtmd.a"
     )
 
     # Create temporary directory for processing
diff --git a/tools/mtmd/CMakeLists.txt b/tools/mtmd/CMakeLists.txt
index 3bafde178de..4f9417c64a2 100644
--- a/tools/mtmd/CMakeLists.txt
+++ b/tools/mtmd/CMakeLists.txt
@@ -95,22 +95,28 @@ if (TARGET mtmd)
     endif()
 endif()
 
-add_executable(llama-llava-cli    deprecation-warning.cpp)
-add_executable(llama-gemma3-cli   deprecation-warning.cpp)
-add_executable(llama-minicpmv-cli deprecation-warning.cpp)
-add_executable(llama-qwen2vl-cli  deprecation-warning.cpp)
+# Gate CLI binaries on LLAMA_BUILD_TOOLS so that standalone library-only
+# builds (LLAMA_BUILD_MTMD=ON with LLAMA_BUILD_TOOLS=OFF — e.g. Apple
+# XCFramework packaging) skip the executables entirely. LLAMA_BUILD_COMMON
+# defaults to ON in standalone builds, so we cannot rely on it for gating.
+if (LLAMA_BUILD_TOOLS)
+    add_executable(llama-llava-cli    deprecation-warning.cpp)
+    add_executable(llama-gemma3-cli   deprecation-warning.cpp)
+    add_executable(llama-minicpmv-cli deprecation-warning.cpp)
+    add_executable(llama-qwen2vl-cli  deprecation-warning.cpp)
 
-set(TARGET llama-mtmd-cli)
-add_executable         (${TARGET} mtmd-cli.cpp)
-set_target_properties  (${TARGET} PROPERTIES OUTPUT_NAME llama-mtmd-cli)
-if(LLAMA_TOOLS_INSTALL)
-    install(TARGETS ${TARGET} RUNTIME)
-endif()
-target_link_libraries  (${TARGET} PRIVATE common mtmd Threads::Threads)
-target_compile_features(${TARGET} PRIVATE cxx_std_17)
+    set(TARGET llama-mtmd-cli)
+    add_executable         (${TARGET} mtmd-cli.cpp)
+    set_target_properties  (${TARGET} PROPERTIES OUTPUT_NAME llama-mtmd-cli)
+    if(LLAMA_TOOLS_INSTALL)
+        install(TARGETS ${TARGET} RUNTIME)
+    endif()
+    target_link_libraries  (${TARGET} PRIVATE common mtmd Threads::Threads)
+    target_compile_features(${TARGET} PRIVATE cxx_std_17)
 
-# mtmd-debug tool
-add_executable(llama-mtmd-debug debug/mtmd-debug.cpp)
-set_target_properties(llama-mtmd-debug PROPERTIES OUTPUT_NAME llama-mtmd-debug)
-target_link_libraries(llama-mtmd-debug PRIVATE common mtmd Threads::Threads)
-target_compile_features(llama-mtmd-debug PRIVATE cxx_std_17)
+    # mtmd-debug tool
+    add_executable(llama-mtmd-debug debug/mtmd-debug.cpp)
+    set_target_properties(llama-mtmd-debug PROPERTIES OUTPUT_NAME llama-mtmd-debug)
+    target_link_libraries(llama-mtmd-debug PRIVATE common mtmd Threads::Threads)
+    target_compile_features(llama-mtmd-debug PRIVATE cxx_std_17)
+endif()