diff --git a/install.sh b/install.sh
index 7e8df1bc64..18aa922f05 100755
--- a/install.sh
+++ b/install.sh
@@ -2460,6 +2460,9 @@ elif case "$TORCH_INDEX_URL" in */rocm*|*/gfx*) true ;; *) false ;; esac; then
     substep "ROCm: $_rocm_root"
     [ -n "$_gpu_rocm_ver" ] && substep "hipconfig: $_gpu_rocm_ver"
     [ -n "$_gpu_disp_mkt" ] && [ -n "$_gpu_disp_gfx" ] && substep "GPU: $_gpu_disp_mkt"
+elif [ "$OS" = "macos" ] && [ "$_ARCH" = "arm64" ]; then
+    # Apple Silicon: PyTorch gets Metal (MPS) acceleration over unified memory, so not CPU-only.
+    step "gpu" "Apple Silicon (Metal, unified memory)"
 else
     step "gpu" "none (CPU-only)" "$C_WARN"
 fi
diff --git a/studio/setup.sh b/studio/setup.sh
index b78029d6e2..75a7fce31e 100755
--- a/studio/setup.sh
+++ b/studio/setup.sh
@@ -968,6 +968,9 @@ elif [ "$_setup_amd_detected" = true ]; then
     substep "ROCm: $_setup_rocm_root"
     [ -n "$_setup_rocm_ver" ] && substep "hipconfig: $_setup_rocm_ver"
     [ -n "$_setup_mkt" ] && [ -n "$_setup_gfx" ] && substep "GPU: $_setup_mkt"
+elif [ "$(uname -s 2>/dev/null)" = "Darwin" ] && [ "$(uname -m 2>/dev/null)" = "arm64" ]; then
+    # Apple Silicon: llama.cpp builds with Metal over unified memory, so not a CPU-only host.
+    step "gpu" "Apple Silicon (Metal, unified memory)"
 else
     step "gpu" "none (chat-only / GGUF)" "$C_WARN"
     substep "Training and GPU inference require an NVIDIA or AMD ROCm GPU."