diff --git a/include/cglm-mat-simd.h b/include/cglm-mat-simd.h
index 65e9f4d..3ca9cc6 100644
--- a/include/cglm-mat-simd.h
+++ b/include/cglm-mat-simd.h
@@ -28,4 +28,24 @@
       _mm_store_ps(D + 12, _mm_madd4_ps(L + 12, r0, r1, r2, r3));             \
    } while (0)
 
+#define CGLM_MAT_TRANSP_SSE_4x4f(M, D)                                        \
+   do {                                                                       \
+      __m128 r0;                                                              \
+      __m128 r1;                                                              \
+      __m128 r2;                                                              \
+      __m128 r3;                                                              \
+                                                                              \
+      r0 = _mm_load_ps(M[0]);                                                 \
+      r1 = _mm_load_ps(M[1]);                                                 \
+      r2 = _mm_load_ps(M[2]);                                                 \
+      r3 = _mm_load_ps(M[3]);                                                 \
+                                                                              \
+      _MM_TRANSPOSE4_PS(r0, r1, r2, r3);                                      \
+                                                                              \
+      _mm_store_ps(D[0], r0);                                                 \
+      _mm_store_ps(D[1], r1);                                                 \
+      _mm_store_ps(D[2], r2);                                                 \
+      _mm_store_ps(D[3], r3);                                                 \
+   } while (0)
+
 #endif /* cglm_mat_sse_h */
diff --git a/include/cglm-mat.h b/include/cglm-mat.h
index aad0a29..81ce791 100644
--- a/include/cglm-mat.h
+++ b/include/cglm-mat.h
@@ -82,4 +82,46 @@ glm_mat_mul4N(mat4 * __restrict matrices[], int len, mat4 dest) {
                  dest);
 }
 
+CGLM_INLINE
+void
+glm_mat_transpose(mat4 m, mat4 dest) {
+#if defined( __SSE__ ) || defined( __SSE2__ )
+  CGLM_MAT_TRANSP_SSE_4x4f(m, dest);
+#else
+  dest[0][0] = m[0][0];
+  dest[0][1] = m[1][0];
+  dest[0][2] = m[2][0];
+  dest[0][3] = m[3][0];
+
+  dest[1][0] = m[0][1];
+  dest[1][1] = m[1][1];
+  dest[1][2] = m[2][1];
+  dest[1][3] = m[3][1];
+
+  dest[2][0] = m[0][2];
+  dest[2][1] = m[1][2];
+  dest[2][2] = m[2][2];
+  dest[2][3] = m[3][2];
+
+  dest[3][0] = m[0][3];
+  dest[3][1] = m[1][3];
+  dest[3][2] = m[2][3];
+  dest[3][3] = m[3][3];
+#endif
+}
+
+CGLM_INLINE
+void
+glm_mat_transpose_self(mat4 m) {
+#if defined( __SSE__ ) || defined( __SSE2__ )
+  CGLM_MAT_TRANSP_SSE_4x4f(m, m);
+#else
+  mat4 d;
+
+  glm_mat_transpose(m, d);
+
+  glm__memcpy(m, d, sizeof(mat4));
+#endif
+}
+
 #endif /* cglm_mat_h */