diff --git a/common/common.cpp b/common/common.cpp
index 0f5fc11a7..90cbe94a3 100644
--- a/common/common.cpp
+++ b/common/common.cpp
@@ -662,9 +662,6 @@ bool gpt_params_parse_ex(int argc, char ** argv, gpt_params & params) {
                /**/ if (value == "interleave" || value == "" )   { params.numa = GGML_NUMA_STRATEGY_INTERLEAVE; }
                else if (value == "isolate") { params.numa = GGML_NUMA_STRATEGY_ISOLATE; }
                else if (value == "numactl")   { params.numa = GGML_NUMA_STRATEGY_NUMACTL; }
-#ifdef GGUF_NUMA_MIRROR
-               else if (value == "mirror")   { params.numa = GGML_NUMA_STRATEGY_MIRROR; }
-#endif
                else { invalid_param = true; break; }
             }
         } else if (arg == "--verbose-prompt") {
@@ -1011,9 +1008,6 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     printf("                          - interleave: (default) spread execution evenly over all nodes\n");
     printf("                          - isolate: only spawn threads on CPUs on the node that execution started on\n");
     printf("                          - numactl: use the CPU map provided my numactl\n");
-#ifdef GGML_NUMA_MIRROR
-    printf("                          - mirror: NOT YET IMPLEMENTED - attempt to mirror GGUF data buffer on each node's local memory to increase throughput.\n");
-#endif
     printf("                        if run without this previously, it is recommended to drop the system page cache before using this\n");
     printf("                        see https://github.com/ggerganov/llama.cpp/issues/1437\n");
     if (llama_supports_gpu_offload()) {
diff --git a/examples/server/server.cpp b/examples/server/server.cpp
index 0c9230d0f..7ed10d564 100644
--- a/examples/server/server.cpp
+++ b/examples/server/server.cpp
@@ -1825,9 +1825,6 @@ static void server_print_usage(const char *argv0, const gpt_params &params,
     printf("                          - interleave: (default) spread execution evenly over all nodes\n");
     printf("                          - isolate: only spawn threads on CPUs on the node that execution started on\n");
     printf("                          - numactl: use the CPU map provided my numactl\n");
-#ifdef GGML_NUMA_MIRROR
-    printf("                          - mirror: NOT YET IMPLEMENTED - attempt to mirror GGUF data buffer on each node's local memory to increase throughput.\n");
-#endif
     if (llama_supports_gpu_offload()) {
         printf("  -ngl N, --n-gpu-layers N\n");
         printf("                            number of layers to store in VRAM\n");
diff --git a/ggml.c b/ggml.c
index 48e156a5b..3d8f20389 100644
--- a/ggml.c
+++ b/ggml.c
@@ -24,10 +24,6 @@
 #include <stdarg.h>
 #include <signal.h>
 
-#ifdef GGML_NUMA_MIRROR
-#include <numa.h>
-#endif
-
 #ifdef GGML_USE_METAL
 #include <unistd.h>
 #endif
@@ -16635,10 +16631,6 @@ static void set_numa_thread_affinity(int thread_n, int n_threads) {
                 fprintf(stderr, "warning: pthread_setaffinity_np() failed: %s\n",strerror(rv));
             }
             return;
-#ifdef GGML_NUMA_MIRROR
-        case GGML_NUMA_STRATEGY_MIRROR:
-            printf("Mirror Mode Enabled");
-#endif
         default:
             return;
     }