numkong 7.0.0 → 7.4.2

package/include/numkong/mesh/skylake.h

@@ -8,11 +8,11 @@
  *
  *  @section skylake_mesh_instructions Key AVX-512 Mesh Instructions
  *
- *      Intrinsic                   Instruction                     Latency     Throughput  Ports
- *      _mm512_fmadd_ps             VFMADD132PS (ZMM, ZMM, ZMM)     4cy         0.5/cy      p05
- *      _mm512_permutexvar_ps       VPERMPS (ZMM, ZMM, ZMM)         3cy         1/cy        p5
- *      _mm512_permutex2var_ps      VPERMT2PS (ZMM, ZMM, ZMM)       3cy         1/cy        p5
- *      _mm512_extractf32x8_ps      VEXTRACTF32X8 (YMM, ZMM, I8)    3cy         1/cy        p5
+ *      Intrinsic               Instruction                   Skylake-X  Genoa
+ *      _mm512_fmadd_ps         VFMADD132PS (ZMM, ZMM, ZMM)   4cy @ p05  4cy @ p01
+ *      _mm512_permutexvar_ps   VPERMPS (ZMM, ZMM, ZMM)       3cy @ p5   4cy @ p12
+ *      _mm512_permutex2var_ps  VPERMT2PS (ZMM, ZMM, ZMM)     3cy @ p5   4cy @ p12
+ *      _mm512_extractf32x8_ps  VEXTRACTF32X8 (YMM, ZMM, I8)  3cy @ p5   1cy @ p0123
  *
  *  Point cloud operations use VPERMT2PS for stride-3 deinterleaving of xyz coordinates, avoiding
  *  expensive gather instructions. This achieves ~1.8x speedup over scalar deinterleaving. Dual FMA
@@ -28,6 +28,7 @@
 #include "numkong/dot/skylake.h"
 #include "numkong/mesh/serial.h"
 #include "numkong/spatial/haswell.h"
+#include "numkong/cast/skylake.h"
 
 #if defined(__cplusplus)
 extern "C" {
@@ -112,6 +113,115 @@ NK_INTERNAL void nk_deinterleave_f64x8_skylake_(
     *z_f64x8_out = _mm512_permutex2var_pd(z01_f64x8, idx_z_2_i64x8, reg2_f64x8);
 }
 
+/*  Deinterleave 16 f16 3D points from xyz,xyz,xyz... to separate x,y,z vectors in f32.
+ *  Input: 48 consecutive f16 values (16 points * 3 coordinates)
+ *  Output: Three __m512 vectors containing the x, y, z coordinates separately (as f32).
+ */
+NK_INTERNAL void nk_deinterleave_f16x16_to_f32x16_skylake_(                                  //
+    nk_f16_t const *ptr, __m512 *x_f32x16_out, __m512 *y_f32x16_out, __m512 *z_f32x16_out) { //
+    __m512 reg0_f32x16 = _mm512_cvtph_ps(_mm256_loadu_si256((__m256i const *)(ptr)));
+    __m512 reg1_f32x16 = _mm512_cvtph_ps(_mm256_loadu_si256((__m256i const *)(ptr + 16)));
+    __m512 reg2_f32x16 = _mm512_cvtph_ps(_mm256_loadu_si256((__m256i const *)(ptr + 32)));
+
+    __m512i idx_x_01_i32x16 = _mm512_setr_epi32(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 0, 0, 0, 0, 0);
+    __m512i idx_x_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 17, 20, 23, 26, 29);
+    __m512 x01_f32x16 = _mm512_permutex2var_ps(reg0_f32x16, idx_x_01_i32x16, reg1_f32x16);
+    *x_f32x16_out = _mm512_permutex2var_ps(x01_f32x16, idx_x_2_i32x16, reg2_f32x16);
+
+    __m512i idx_y_01_i32x16 = _mm512_setr_epi32(1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 0, 0, 0, 0, 0);
+    __m512i idx_y_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 18, 21, 24, 27, 30);
+    __m512 y01_f32x16 = _mm512_permutex2var_ps(reg0_f32x16, idx_y_01_i32x16, reg1_f32x16);
+    *y_f32x16_out = _mm512_permutex2var_ps(y01_f32x16, idx_y_2_i32x16, reg2_f32x16);
+
+    __m512i idx_z_01_i32x16 = _mm512_setr_epi32(2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 0, 0, 0, 0, 0, 0);
+    __m512i idx_z_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 16, 19, 22, 25, 28, 31);
+    __m512 z01_f32x16 = _mm512_permutex2var_ps(reg0_f32x16, idx_z_01_i32x16, reg1_f32x16);
+    *z_f32x16_out = _mm512_permutex2var_ps(z01_f32x16, idx_z_2_i32x16, reg2_f32x16);
+}
+
+/*  Deinterleave 16 bf16 3D points from xyz,xyz,xyz... to separate x,y,z vectors in f32.
+ *  Input: 48 consecutive bf16 values (16 points * 3 coordinates)
+ *  Output: Three __m512 vectors containing the x, y, z coordinates separately (as f32).
+ */
+NK_INTERNAL void nk_deinterleave_bf16x16_to_f32x16_skylake_(                                  //
+    nk_bf16_t const *ptr, __m512 *x_f32x16_out, __m512 *y_f32x16_out, __m512 *z_f32x16_out) { //
+    __m512 reg0_f32x16 = nk_bf16x16_to_f32x16_skylake_(_mm256_loadu_si256((__m256i const *)(ptr)));
+    __m512 reg1_f32x16 = nk_bf16x16_to_f32x16_skylake_(_mm256_loadu_si256((__m256i const *)(ptr + 16)));
+    __m512 reg2_f32x16 = nk_bf16x16_to_f32x16_skylake_(_mm256_loadu_si256((__m256i const *)(ptr + 32)));
+
+    __m512i idx_x_01_i32x16 = _mm512_setr_epi32(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 0, 0, 0, 0, 0);
+    __m512i idx_x_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 17, 20, 23, 26, 29);
+    __m512 x01_f32x16 = _mm512_permutex2var_ps(reg0_f32x16, idx_x_01_i32x16, reg1_f32x16);
+    *x_f32x16_out = _mm512_permutex2var_ps(x01_f32x16, idx_x_2_i32x16, reg2_f32x16);
+
+    __m512i idx_y_01_i32x16 = _mm512_setr_epi32(1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 0, 0, 0, 0, 0);
+    __m512i idx_y_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 18, 21, 24, 27, 30);
+    __m512 y01_f32x16 = _mm512_permutex2var_ps(reg0_f32x16, idx_y_01_i32x16, reg1_f32x16);
+    *y_f32x16_out = _mm512_permutex2var_ps(y01_f32x16, idx_y_2_i32x16, reg2_f32x16);
+
+    __m512i idx_z_01_i32x16 = _mm512_setr_epi32(2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 0, 0, 0, 0, 0, 0);
+    __m512i idx_z_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 16, 19, 22, 25, 28, 31);
+    __m512 z01_f32x16 = _mm512_permutex2var_ps(reg0_f32x16, idx_z_01_i32x16, reg1_f32x16);
+    *z_f32x16_out = _mm512_permutex2var_ps(z01_f32x16, idx_z_2_i32x16, reg2_f32x16);
+}
+
+/*  Masked-tail deinterleave for f16: loads up to 16 xyz points using AVX-512 masked loads,
+ *  converts f16→f32, and deinterleaves into separate x,y,z vectors.
+ *  Unused lanes are zero. Uses the same permutex2var shuffle as the full-width version.
+ */
+NK_INTERNAL void nk_deinterleave_f16_tail_to_f32x16_skylake_(                                                 //
+    nk_f16_t const *ptr, nk_size_t count, __m512 *x_f32x16_out, __m512 *y_f32x16_out, __m512 *z_f32x16_out) { //
+    nk_size_t total = count * 3;
+    __mmask16 mask0_i16x16 = (__mmask16)_bzhi_u32(0xFFFF, total >= 16 ? 16 : total);
+    __mmask16 mask1_i16x16 = total > 16 ? (__mmask16)_bzhi_u32(0xFFFF, total >= 32 ? 16 : total - 16) : 0;
+    __mmask16 mask2_i16x16 = total > 32 ? (__mmask16)_bzhi_u32(0xFFFF, total - 32) : 0;
+    __m512 reg0_f32x16 = _mm512_cvtph_ps(_mm256_maskz_loadu_epi16(mask0_i16x16, ptr));
+    __m512 reg1_f32x16 = _mm512_cvtph_ps(_mm256_maskz_loadu_epi16(mask1_i16x16, ptr + 16));
+    __m512 reg2_f32x16 = _mm512_cvtph_ps(_mm256_maskz_loadu_epi16(mask2_i16x16, ptr + 32));
+
+    __m512i idx_x_01_i32x16 = _mm512_setr_epi32(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 0, 0, 0, 0, 0);
+    __m512i idx_x_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 17, 20, 23, 26, 29);
+    *x_f32x16_out = _mm512_permutex2var_ps(_mm512_permutex2var_ps(reg0_f32x16, idx_x_01_i32x16, reg1_f32x16),
+                                           idx_x_2_i32x16, reg2_f32x16);
+
+    __m512i idx_y_01_i32x16 = _mm512_setr_epi32(1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 0, 0, 0, 0, 0);
+    __m512i idx_y_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 18, 21, 24, 27, 30);
+    *y_f32x16_out = _mm512_permutex2var_ps(_mm512_permutex2var_ps(reg0_f32x16, idx_y_01_i32x16, reg1_f32x16),
+                                           idx_y_2_i32x16, reg2_f32x16);
+
+    __m512i idx_z_01_i32x16 = _mm512_setr_epi32(2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 0, 0, 0, 0, 0, 0);
+    __m512i idx_z_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 16, 19, 22, 25, 28, 31);
+    *z_f32x16_out = _mm512_permutex2var_ps(_mm512_permutex2var_ps(reg0_f32x16, idx_z_01_i32x16, reg1_f32x16),
+                                           idx_z_2_i32x16, reg2_f32x16);
+}
+
+/*  Masked-tail deinterleave for bf16: same as f16 but with bf16→f32 conversion. */
+NK_INTERNAL void nk_deinterleave_bf16_tail_to_f32x16_skylake_(                                                 //
+    nk_bf16_t const *ptr, nk_size_t count, __m512 *x_f32x16_out, __m512 *y_f32x16_out, __m512 *z_f32x16_out) { //
+    nk_size_t total = count * 3;
+    __mmask16 mask0_i16x16 = (__mmask16)_bzhi_u32(0xFFFF, total >= 16 ? 16 : total);
+    __mmask16 mask1_i16x16 = total > 16 ? (__mmask16)_bzhi_u32(0xFFFF, total >= 32 ? 16 : total - 16) : 0;
+    __mmask16 mask2_i16x16 = total > 32 ? (__mmask16)_bzhi_u32(0xFFFF, total - 32) : 0;
+    __m512 reg0_f32x16 = nk_bf16x16_to_f32x16_skylake_(_mm256_maskz_loadu_epi16(mask0_i16x16, ptr));
+    __m512 reg1_f32x16 = nk_bf16x16_to_f32x16_skylake_(_mm256_maskz_loadu_epi16(mask1_i16x16, ptr + 16));
+    __m512 reg2_f32x16 = nk_bf16x16_to_f32x16_skylake_(_mm256_maskz_loadu_epi16(mask2_i16x16, ptr + 32));
+
+    __m512i idx_x_01_i32x16 = _mm512_setr_epi32(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 0, 0, 0, 0, 0);
+    __m512i idx_x_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 17, 20, 23, 26, 29);
+    *x_f32x16_out = _mm512_permutex2var_ps(_mm512_permutex2var_ps(reg0_f32x16, idx_x_01_i32x16, reg1_f32x16),
+                                           idx_x_2_i32x16, reg2_f32x16);
+
+    __m512i idx_y_01_i32x16 = _mm512_setr_epi32(1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 0, 0, 0, 0, 0);
+    __m512i idx_y_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 18, 21, 24, 27, 30);
+    *y_f32x16_out = _mm512_permutex2var_ps(_mm512_permutex2var_ps(reg0_f32x16, idx_y_01_i32x16, reg1_f32x16),
+                                           idx_y_2_i32x16, reg2_f32x16);
+
+    __m512i idx_z_01_i32x16 = _mm512_setr_epi32(2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 0, 0, 0, 0, 0, 0);
+    __m512i idx_z_2_i32x16 = _mm512_setr_epi32(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 16, 19, 22, 25, 28, 31);
+    *z_f32x16_out = _mm512_permutex2var_ps(_mm512_permutex2var_ps(reg0_f32x16, idx_z_01_i32x16, reg1_f32x16),
+                                           idx_z_2_i32x16, reg2_f32x16);
+}
+
 NK_INTERNAL nk_f64_t nk_reduce_stable_f64x8_skylake_(__m512d values_f64x8) {
     nk_b512_vec_t values;
     values.zmm_pd = values_f64x8;
@@ -166,84 +276,84 @@ NK_INTERNAL nk_f64_t nk_transformed_ssd_f32_skylake_(nk_f32_t const *a, nk_f32_t
     for (; index + 16 <= n; index += 16) {
         nk_deinterleave_f32x16_skylake_(a + index * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16),
             nk_deinterleave_f32x16_skylake_(b + index * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
-        __m512d a_x_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
-        __m512d a_x_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
-        __m512d a_y_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
-        __m512d a_y_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
-        __m512d a_z_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
-        __m512d a_z_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
-        __m512d b_x_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
-        __m512d b_x_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
-        __m512d b_y_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
-        __m512d b_y_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
-        __m512d b_z_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
-        __m512d b_z_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
-
-        __m512d centered_a_x_lower_f64x8 = _mm512_sub_pd(a_x_lower_f64x8, centroid_a_x_f64x8);
-        __m512d centered_a_x_upper_f64x8 = _mm512_sub_pd(a_x_upper_f64x8, centroid_a_x_f64x8);
-        __m512d centered_a_y_lower_f64x8 = _mm512_sub_pd(a_y_lower_f64x8, centroid_a_y_f64x8);
-        __m512d centered_a_y_upper_f64x8 = _mm512_sub_pd(a_y_upper_f64x8, centroid_a_y_f64x8);
-        __m512d centered_a_z_lower_f64x8 = _mm512_sub_pd(a_z_lower_f64x8, centroid_a_z_f64x8);
-        __m512d centered_a_z_upper_f64x8 = _mm512_sub_pd(a_z_upper_f64x8, centroid_a_z_f64x8);
-        __m512d centered_b_x_lower_f64x8 = _mm512_sub_pd(b_x_lower_f64x8, centroid_b_x_f64x8);
-        __m512d centered_b_x_upper_f64x8 = _mm512_sub_pd(b_x_upper_f64x8, centroid_b_x_f64x8);
-        __m512d centered_b_y_lower_f64x8 = _mm512_sub_pd(b_y_lower_f64x8, centroid_b_y_f64x8);
-        __m512d centered_b_y_upper_f64x8 = _mm512_sub_pd(b_y_upper_f64x8, centroid_b_y_f64x8);
-        __m512d centered_b_z_lower_f64x8 = _mm512_sub_pd(b_z_lower_f64x8, centroid_b_z_f64x8);
-        __m512d centered_b_z_upper_f64x8 = _mm512_sub_pd(b_z_upper_f64x8, centroid_b_z_f64x8);
-
-        __m512d rotated_a_x_lower_f64x8 = _mm512_fmadd_pd(
-            scaled_rotation_x_z_f64x8, centered_a_z_lower_f64x8,
-            _mm512_fmadd_pd(scaled_rotation_x_y_f64x8, centered_a_y_lower_f64x8,
-                            _mm512_mul_pd(scaled_rotation_x_x_f64x8, centered_a_x_lower_f64x8)));
-        __m512d rotated_a_x_upper_f64x8 = _mm512_fmadd_pd(
-            scaled_rotation_x_z_f64x8, centered_a_z_upper_f64x8,
-            _mm512_fmadd_pd(scaled_rotation_x_y_f64x8, centered_a_y_upper_f64x8,
-                            _mm512_mul_pd(scaled_rotation_x_x_f64x8, centered_a_x_upper_f64x8)));
-        __m512d rotated_a_y_lower_f64x8 = _mm512_fmadd_pd(
-            scaled_rotation_y_z_f64x8, centered_a_z_lower_f64x8,
-            _mm512_fmadd_pd(scaled_rotation_y_y_f64x8, centered_a_y_lower_f64x8,
-                            _mm512_mul_pd(scaled_rotation_y_x_f64x8, centered_a_x_lower_f64x8)));
-        __m512d rotated_a_y_upper_f64x8 = _mm512_fmadd_pd(
-            scaled_rotation_y_z_f64x8, centered_a_z_upper_f64x8,
-            _mm512_fmadd_pd(scaled_rotation_y_y_f64x8, centered_a_y_upper_f64x8,
-                            _mm512_mul_pd(scaled_rotation_y_x_f64x8, centered_a_x_upper_f64x8)));
-        __m512d rotated_a_z_lower_f64x8 = _mm512_fmadd_pd(
-            scaled_rotation_z_z_f64x8, centered_a_z_lower_f64x8,
-            _mm512_fmadd_pd(scaled_rotation_z_y_f64x8, centered_a_y_lower_f64x8,
-                            _mm512_mul_pd(scaled_rotation_z_x_f64x8, centered_a_x_lower_f64x8)));
-        __m512d rotated_a_z_upper_f64x8 = _mm512_fmadd_pd(
-            scaled_rotation_z_z_f64x8, centered_a_z_upper_f64x8,
-            _mm512_fmadd_pd(scaled_rotation_z_y_f64x8, centered_a_y_upper_f64x8,
-                            _mm512_mul_pd(scaled_rotation_z_x_f64x8, centered_a_x_upper_f64x8)));
-
-        __m512d delta_x_lower_f64x8 = _mm512_sub_pd(rotated_a_x_lower_f64x8, centered_b_x_lower_f64x8);
-        __m512d delta_x_upper_f64x8 = _mm512_sub_pd(rotated_a_x_upper_f64x8, centered_b_x_upper_f64x8);
-        __m512d delta_y_lower_f64x8 = _mm512_sub_pd(rotated_a_y_lower_f64x8, centered_b_y_lower_f64x8);
-        __m512d delta_y_upper_f64x8 = _mm512_sub_pd(rotated_a_y_upper_f64x8, centered_b_y_upper_f64x8);
-        __m512d delta_z_lower_f64x8 = _mm512_sub_pd(rotated_a_z_lower_f64x8, centered_b_z_lower_f64x8);
-        __m512d delta_z_upper_f64x8 = _mm512_sub_pd(rotated_a_z_upper_f64x8, centered_b_z_upper_f64x8);
-
-        __m512d batch_sum_squared_f64x8 = _mm512_add_pd(_mm512_mul_pd(delta_x_lower_f64x8, delta_x_lower_f64x8),
-                                                        _mm512_mul_pd(delta_x_upper_f64x8, delta_x_upper_f64x8));
-        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_y_lower_f64x8, delta_y_lower_f64x8, batch_sum_squared_f64x8);
-        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_y_upper_f64x8, delta_y_upper_f64x8, batch_sum_squared_f64x8);
-        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_z_lower_f64x8, delta_z_lower_f64x8, batch_sum_squared_f64x8);
-        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_z_upper_f64x8, delta_z_upper_f64x8, batch_sum_squared_f64x8);
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        __m512d centered_a_x_low_f64x8 = _mm512_sub_pd(a_x_low_f64x8, centroid_a_x_f64x8);
+        __m512d centered_a_x_high_f64x8 = _mm512_sub_pd(a_x_high_f64x8, centroid_a_x_f64x8);
+        __m512d centered_a_y_low_f64x8 = _mm512_sub_pd(a_y_low_f64x8, centroid_a_y_f64x8);
+        __m512d centered_a_y_high_f64x8 = _mm512_sub_pd(a_y_high_f64x8, centroid_a_y_f64x8);
+        __m512d centered_a_z_low_f64x8 = _mm512_sub_pd(a_z_low_f64x8, centroid_a_z_f64x8);
+        __m512d centered_a_z_high_f64x8 = _mm512_sub_pd(a_z_high_f64x8, centroid_a_z_f64x8);
+        __m512d centered_b_x_low_f64x8 = _mm512_sub_pd(b_x_low_f64x8, centroid_b_x_f64x8);
+        __m512d centered_b_x_high_f64x8 = _mm512_sub_pd(b_x_high_f64x8, centroid_b_x_f64x8);
+        __m512d centered_b_y_low_f64x8 = _mm512_sub_pd(b_y_low_f64x8, centroid_b_y_f64x8);
+        __m512d centered_b_y_high_f64x8 = _mm512_sub_pd(b_y_high_f64x8, centroid_b_y_f64x8);
+        __m512d centered_b_z_low_f64x8 = _mm512_sub_pd(b_z_low_f64x8, centroid_b_z_f64x8);
+        __m512d centered_b_z_high_f64x8 = _mm512_sub_pd(b_z_high_f64x8, centroid_b_z_f64x8);
+
+        __m512d rotated_a_x_low_f64x8 = _mm512_fmadd_pd(
+            scaled_rotation_x_z_f64x8, centered_a_z_low_f64x8,
+            _mm512_fmadd_pd(scaled_rotation_x_y_f64x8, centered_a_y_low_f64x8,
+                            _mm512_mul_pd(scaled_rotation_x_x_f64x8, centered_a_x_low_f64x8)));
+        __m512d rotated_a_x_high_f64x8 = _mm512_fmadd_pd(
+            scaled_rotation_x_z_f64x8, centered_a_z_high_f64x8,
+            _mm512_fmadd_pd(scaled_rotation_x_y_f64x8, centered_a_y_high_f64x8,
+                            _mm512_mul_pd(scaled_rotation_x_x_f64x8, centered_a_x_high_f64x8)));
+        __m512d rotated_a_y_low_f64x8 = _mm512_fmadd_pd(
+            scaled_rotation_y_z_f64x8, centered_a_z_low_f64x8,
+            _mm512_fmadd_pd(scaled_rotation_y_y_f64x8, centered_a_y_low_f64x8,
+                            _mm512_mul_pd(scaled_rotation_y_x_f64x8, centered_a_x_low_f64x8)));
+        __m512d rotated_a_y_high_f64x8 = _mm512_fmadd_pd(
+            scaled_rotation_y_z_f64x8, centered_a_z_high_f64x8,
+            _mm512_fmadd_pd(scaled_rotation_y_y_f64x8, centered_a_y_high_f64x8,
+                            _mm512_mul_pd(scaled_rotation_y_x_f64x8, centered_a_x_high_f64x8)));
+        __m512d rotated_a_z_low_f64x8 = _mm512_fmadd_pd(
+            scaled_rotation_z_z_f64x8, centered_a_z_low_f64x8,
+            _mm512_fmadd_pd(scaled_rotation_z_y_f64x8, centered_a_y_low_f64x8,
+                            _mm512_mul_pd(scaled_rotation_z_x_f64x8, centered_a_x_low_f64x8)));
+        __m512d rotated_a_z_high_f64x8 = _mm512_fmadd_pd(
+            scaled_rotation_z_z_f64x8, centered_a_z_high_f64x8,
+            _mm512_fmadd_pd(scaled_rotation_z_y_f64x8, centered_a_y_high_f64x8,
+                            _mm512_mul_pd(scaled_rotation_z_x_f64x8, centered_a_x_high_f64x8)));
+
+        __m512d delta_x_low_f64x8 = _mm512_sub_pd(rotated_a_x_low_f64x8, centered_b_x_low_f64x8);
+        __m512d delta_x_high_f64x8 = _mm512_sub_pd(rotated_a_x_high_f64x8, centered_b_x_high_f64x8);
+        __m512d delta_y_low_f64x8 = _mm512_sub_pd(rotated_a_y_low_f64x8, centered_b_y_low_f64x8);
+        __m512d delta_y_high_f64x8 = _mm512_sub_pd(rotated_a_y_high_f64x8, centered_b_y_high_f64x8);
+        __m512d delta_z_low_f64x8 = _mm512_sub_pd(rotated_a_z_low_f64x8, centered_b_z_low_f64x8);
+        __m512d delta_z_high_f64x8 = _mm512_sub_pd(rotated_a_z_high_f64x8, centered_b_z_high_f64x8);
+
+        __m512d batch_sum_squared_f64x8 = _mm512_add_pd(_mm512_mul_pd(delta_x_low_f64x8, delta_x_low_f64x8),
+                                                        _mm512_mul_pd(delta_x_high_f64x8, delta_x_high_f64x8));
+        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_y_low_f64x8, delta_y_low_f64x8, batch_sum_squared_f64x8);
+        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_y_high_f64x8, delta_y_high_f64x8, batch_sum_squared_f64x8);
+        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_z_low_f64x8, delta_z_low_f64x8, batch_sum_squared_f64x8);
+        batch_sum_squared_f64x8 = _mm512_fmadd_pd(delta_z_high_f64x8, delta_z_high_f64x8, batch_sum_squared_f64x8);
         sum_squared_f64x8 = _mm512_add_pd(sum_squared_f64x8, batch_sum_squared_f64x8);
     }
 
     nk_f64_t sum_squared = _mm512_reduce_add_pd(sum_squared_f64x8);
     for (; index < n; ++index) {
-        nk_f64_t centered_a_x = (nk_f64_t)a[index * 3 + 0] - centroid_a_x;
-        nk_f64_t centered_a_y = (nk_f64_t)a[index * 3 + 1] - centroid_a_y;
-        nk_f64_t centered_a_z = (nk_f64_t)a[index * 3 + 2] - centroid_a_z;
-        nk_f64_t centered_b_x = (nk_f64_t)b[index * 3 + 0] - centroid_b_x;
-        nk_f64_t centered_b_y = (nk_f64_t)b[index * 3 + 1] - centroid_b_y;
-        nk_f64_t centered_b_z = (nk_f64_t)b[index * 3 + 2] - centroid_b_z;
-        nk_f64_t rotated_a_x = scale * (r[0] * centered_a_x + r[1] * centered_a_y + r[2] * centered_a_z);
-        nk_f64_t rotated_a_y = scale * (r[3] * centered_a_x + r[4] * centered_a_y + r[5] * centered_a_z);
-        nk_f64_t rotated_a_z = scale * (r[6] * centered_a_x + r[7] * centered_a_y + r[8] * centered_a_z);
+        nk_f64_t centered_a_x = (nk_f64_t)a[index * 3 + 0] - centroid_a_x,
+                 centered_a_y = (nk_f64_t)a[index * 3 + 1] - centroid_a_y,
+                 centered_a_z = (nk_f64_t)a[index * 3 + 2] - centroid_a_z;
+        nk_f64_t centered_b_x = (nk_f64_t)b[index * 3 + 0] - centroid_b_x,
+                 centered_b_y = (nk_f64_t)b[index * 3 + 1] - centroid_b_y,
+                 centered_b_z = (nk_f64_t)b[index * 3 + 2] - centroid_b_z;
+        nk_f64_t rotated_a_x = scale * (r[0] * centered_a_x + r[1] * centered_a_y + r[2] * centered_a_z),
+                 rotated_a_y = scale * (r[3] * centered_a_x + r[4] * centered_a_y + r[5] * centered_a_z),
+                 rotated_a_z = scale * (r[6] * centered_a_x + r[7] * centered_a_y + r[8] * centered_a_z);
         nk_f64_t delta_x = rotated_a_x - centered_b_x, delta_y = rotated_a_y - centered_b_y,
                  delta_z = rotated_a_z - centered_b_z;
         sum_squared += delta_x * delta_x + delta_y * delta_y + delta_z * delta_z;
@@ -322,20 +432,16 @@ NK_INTERNAL nk_f64_t nk_transformed_ssd_f64_skylake_(nk_f64_t const *a, nk_f64_t
 
     // Scalar tail
     for (; j < n; ++j) {
-        nk_f64_t pa_x = a[j * 3 + 0] - centroid_a_x;
-        nk_f64_t pa_y = a[j * 3 + 1] - centroid_a_y;
-        nk_f64_t pa_z = a[j * 3 + 2] - centroid_a_z;
-        nk_f64_t pb_x = b[j * 3 + 0] - centroid_b_x;
-        nk_f64_t pb_y = b[j * 3 + 1] - centroid_b_y;
-        nk_f64_t pb_z = b[j * 3 + 2] - centroid_b_z;
-
-        nk_f64_t ra_x = scale * (r[0] * pa_x + r[1] * pa_y + r[2] * pa_z);
-        nk_f64_t ra_y = scale * (r[3] * pa_x + r[4] * pa_y + r[5] * pa_z);
-        nk_f64_t ra_z = scale * (r[6] * pa_x + r[7] * pa_y + r[8] * pa_z);
-
-        nk_f64_t delta_x = ra_x - pb_x;
-        nk_f64_t delta_y = ra_y - pb_y;
-        nk_f64_t delta_z = ra_z - pb_z;
+        nk_f64_t pa_x = a[j * 3 + 0] - centroid_a_x, pa_y = a[j * 3 + 1] - centroid_a_y,
+                 pa_z = a[j * 3 + 2] - centroid_a_z;
+        nk_f64_t pb_x = b[j * 3 + 0] - centroid_b_x, pb_y = b[j * 3 + 1] - centroid_b_y,
+                 pb_z = b[j * 3 + 2] - centroid_b_z;
+
+        nk_f64_t ra_x = scale * (r[0] * pa_x + r[1] * pa_y + r[2] * pa_z),
+                 ra_y = scale * (r[3] * pa_x + r[4] * pa_y + r[5] * pa_z),
+                 ra_z = scale * (r[6] * pa_x + r[7] * pa_y + r[8] * pa_z);
+
+        nk_f64_t delta_x = ra_x - pb_x, delta_y = ra_y - pb_y, delta_z = ra_z - pb_z;
         nk_accumulate_square_f64_(&sum_squared, &sum_squared_compensation, delta_x);
         nk_accumulate_square_f64_(&sum_squared, &sum_squared_compensation, delta_y);
         nk_accumulate_square_f64_(&sum_squared, &sum_squared_compensation, delta_z);
@@ -344,139 +450,526 @@ NK_INTERNAL nk_f64_t nk_transformed_ssd_f64_skylake_(nk_f64_t const *a, nk_f64_t
     return sum_squared + sum_squared_compensation;
 }
 
-NK_INTERNAL void nk_centroid_and_cross_covariance_f32_skylake_( //
-    nk_f32_t const *a, nk_f32_t const *b, nk_size_t n,          //
-    nk_f64_t *centroid_a_x, nk_f64_t *centroid_a_y, nk_f64_t *centroid_a_z, nk_f64_t *centroid_b_x,
-    nk_f64_t *centroid_b_y, nk_f64_t *centroid_b_z, nk_f64_t cross_covariance_f64[9]) {
-    __m512d sum_a_x_f64x8 = _mm512_setzero_pd(), sum_a_y_f64x8 = _mm512_setzero_pd();
-    __m512d sum_a_z_f64x8 = _mm512_setzero_pd(), sum_b_x_f64x8 = _mm512_setzero_pd();
-    __m512d sum_b_y_f64x8 = _mm512_setzero_pd(), sum_b_z_f64x8 = _mm512_setzero_pd();
-    __m512d covariance_00_f64x8 = _mm512_setzero_pd(), covariance_01_f64x8 = _mm512_setzero_pd();
-    __m512d covariance_02_f64x8 = _mm512_setzero_pd(), covariance_10_f64x8 = _mm512_setzero_pd();
-    __m512d covariance_11_f64x8 = _mm512_setzero_pd(), covariance_12_f64x8 = _mm512_setzero_pd();
-    __m512d covariance_20_f64x8 = _mm512_setzero_pd(), covariance_21_f64x8 = _mm512_setzero_pd();
-    __m512d covariance_22_f64x8 = _mm512_setzero_pd();
+/*  Compute sum of squared distances for f16 data after applying rotation (and optional scale).
+ *  Loads f16, converts to f32 for computation. Rotation matrix, scale, and centroids are f32.
+ */
+NK_INTERNAL nk_f32_t nk_transformed_ssd_f16_skylake_(nk_f16_t const *a, nk_f16_t const *b, nk_size_t n,
+                                                     nk_f32_t const *r, nk_f32_t scale, nk_f32_t centroid_a_x,
+                                                     nk_f32_t centroid_a_y, nk_f32_t centroid_a_z,
+                                                     nk_f32_t centroid_b_x, nk_f32_t centroid_b_y,
+                                                     nk_f32_t centroid_b_z) {
+    __m512 scaled_rotation_x_x_f32x16 = _mm512_set1_ps(scale * r[0]);
+    __m512 scaled_rotation_x_y_f32x16 = _mm512_set1_ps(scale * r[1]);
+    __m512 scaled_rotation_x_z_f32x16 = _mm512_set1_ps(scale * r[2]);
+    __m512 scaled_rotation_y_x_f32x16 = _mm512_set1_ps(scale * r[3]);
+    __m512 scaled_rotation_y_y_f32x16 = _mm512_set1_ps(scale * r[4]);
+    __m512 scaled_rotation_y_z_f32x16 = _mm512_set1_ps(scale * r[5]);
+    __m512 scaled_rotation_z_x_f32x16 = _mm512_set1_ps(scale * r[6]);
+    __m512 scaled_rotation_z_y_f32x16 = _mm512_set1_ps(scale * r[7]);
+    __m512 scaled_rotation_z_z_f32x16 = _mm512_set1_ps(scale * r[8]);
+
+    __m512 centroid_a_x_f32x16 = _mm512_set1_ps(centroid_a_x);
+    __m512 centroid_a_y_f32x16 = _mm512_set1_ps(centroid_a_y);
+    __m512 centroid_a_z_f32x16 = _mm512_set1_ps(centroid_a_z);
+    __m512 centroid_b_x_f32x16 = _mm512_set1_ps(centroid_b_x);
+    __m512 centroid_b_y_f32x16 = _mm512_set1_ps(centroid_b_y);
+    __m512 centroid_b_z_f32x16 = _mm512_set1_ps(centroid_b_z);
+
+    __m512 sum_squared_f32x16 = _mm512_setzero_ps();
     __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
-    nk_size_t index = 0;
+    nk_size_t j = 0;
 
-    for (; index + 16 <= n; index += 16) {
-        nk_deinterleave_f32x16_skylake_(a + index * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16),
-            nk_deinterleave_f32x16_skylake_(b + index * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
-        __m512d a_x_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
-        __m512d a_x_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
-        __m512d a_y_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
-        __m512d a_y_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
-        __m512d a_z_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
-        __m512d a_z_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
-        __m512d b_x_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
-        __m512d b_x_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
-        __m512d b_y_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
-        __m512d b_y_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
-        __m512d b_z_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
-        __m512d b_z_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
-
-        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_lower_f64x8, a_x_upper_f64x8)),
-        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_lower_f64x8, a_y_upper_f64x8)),
-        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_lower_f64x8, a_z_upper_f64x8));
-        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_lower_f64x8, b_x_upper_f64x8)),
-        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_lower_f64x8, b_y_upper_f64x8)),
-        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_lower_f64x8, b_z_upper_f64x8));
-        covariance_00_f64x8 = _mm512_add_pd(covariance_00_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_x_lower_f64x8, b_x_lower_f64x8),
-                                                          _mm512_mul_pd(a_x_upper_f64x8, b_x_upper_f64x8))),
-        covariance_01_f64x8 = _mm512_add_pd(covariance_01_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_x_lower_f64x8, b_y_lower_f64x8),
-                                                          _mm512_mul_pd(a_x_upper_f64x8, b_y_upper_f64x8))),
-        covariance_02_f64x8 = _mm512_add_pd(covariance_02_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_x_lower_f64x8, b_z_lower_f64x8),
-                                                          _mm512_mul_pd(a_x_upper_f64x8, b_z_upper_f64x8)));
-        covariance_10_f64x8 = _mm512_add_pd(covariance_10_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_y_lower_f64x8, b_x_lower_f64x8),
-                                                          _mm512_mul_pd(a_y_upper_f64x8, b_x_upper_f64x8))),
-        covariance_11_f64x8 = _mm512_add_pd(covariance_11_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_y_lower_f64x8, b_y_lower_f64x8),
-                                                          _mm512_mul_pd(a_y_upper_f64x8, b_y_upper_f64x8))),
-        covariance_12_f64x8 = _mm512_add_pd(covariance_12_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_y_lower_f64x8, b_z_lower_f64x8),
-                                                          _mm512_mul_pd(a_y_upper_f64x8, b_z_upper_f64x8)));
-        covariance_20_f64x8 = _mm512_add_pd(covariance_20_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_z_lower_f64x8, b_x_lower_f64x8),
-                                                          _mm512_mul_pd(a_z_upper_f64x8, b_x_upper_f64x8))),
-        covariance_21_f64x8 = _mm512_add_pd(covariance_21_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_z_lower_f64x8, b_y_lower_f64x8),
-                                                          _mm512_mul_pd(a_z_upper_f64x8, b_y_upper_f64x8))),
-        covariance_22_f64x8 = _mm512_add_pd(covariance_22_f64x8,
-                                            _mm512_add_pd(_mm512_mul_pd(a_z_lower_f64x8, b_z_lower_f64x8),
-                                                          _mm512_mul_pd(a_z_upper_f64x8, b_z_upper_f64x8)));
+    for (; j + 16 <= n; j += 16) {
+        nk_deinterleave_f16x16_to_f32x16_skylake_(a + j * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16x16_to_f32x16_skylake_(b + j * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        __m512 pa_x_f32x16 = _mm512_sub_ps(a_x_f32x16, centroid_a_x_f32x16);
+        __m512 pa_y_f32x16 = _mm512_sub_ps(a_y_f32x16, centroid_a_y_f32x16);
+        __m512 pa_z_f32x16 = _mm512_sub_ps(a_z_f32x16, centroid_a_z_f32x16);
+        __m512 pb_x_f32x16 = _mm512_sub_ps(b_x_f32x16, centroid_b_x_f32x16);
+        __m512 pb_y_f32x16 = _mm512_sub_ps(b_y_f32x16, centroid_b_y_f32x16);
+        __m512 pb_z_f32x16 = _mm512_sub_ps(b_z_f32x16, centroid_b_z_f32x16);
+
+        __m512 ra_x_f32x16 = _mm512_fmadd_ps(scaled_rotation_x_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_x_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_x_x_f32x16, pa_x_f32x16)));
+        __m512 ra_y_f32x16 = _mm512_fmadd_ps(scaled_rotation_y_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_y_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_y_x_f32x16, pa_x_f32x16)));
+        __m512 ra_z_f32x16 = _mm512_fmadd_ps(scaled_rotation_z_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_z_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_z_x_f32x16, pa_x_f32x16)));
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(ra_x_f32x16, pb_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(ra_y_f32x16, pb_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(ra_z_f32x16, pb_z_f32x16);
+
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_f32x16);
     }
 
-    nk_f64_t sum_a_x = _mm512_reduce_add_pd(sum_a_x_f64x8), sum_a_y = _mm512_reduce_add_pd(sum_a_y_f64x8),
-             sum_a_z = _mm512_reduce_add_pd(sum_a_z_f64x8);
-    nk_f64_t sum_b_x = _mm512_reduce_add_pd(sum_b_x_f64x8), sum_b_y = _mm512_reduce_add_pd(sum_b_y_f64x8),
-             sum_b_z = _mm512_reduce_add_pd(sum_b_z_f64x8);
-    nk_f64_t covariance_00 = _mm512_reduce_add_pd(covariance_00_f64x8),
-             covariance_01 = _mm512_reduce_add_pd(covariance_01_f64x8),
-             covariance_02 = _mm512_reduce_add_pd(covariance_02_f64x8);
-    nk_f64_t covariance_10 = _mm512_reduce_add_pd(covariance_10_f64x8),
-             covariance_11 = _mm512_reduce_add_pd(covariance_11_f64x8),
-             covariance_12 = _mm512_reduce_add_pd(covariance_12_f64x8);
-    nk_f64_t covariance_20 = _mm512_reduce_add_pd(covariance_20_f64x8),
-             covariance_21 = _mm512_reduce_add_pd(covariance_21_f64x8),
-             covariance_22 = _mm512_reduce_add_pd(covariance_22_f64x8);
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (j < n) {
+        nk_size_t tail = n - j;
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(a + j * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(b + j * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        __m512 pa_x_f32x16 = _mm512_sub_ps(a_x_f32x16, centroid_a_x_f32x16);
+        __m512 pa_y_f32x16 = _mm512_sub_ps(a_y_f32x16, centroid_a_y_f32x16);
+        __m512 pa_z_f32x16 = _mm512_sub_ps(a_z_f32x16, centroid_a_z_f32x16);
+        __m512 pb_x_f32x16 = _mm512_sub_ps(b_x_f32x16, centroid_b_x_f32x16);
+        __m512 pb_y_f32x16 = _mm512_sub_ps(b_y_f32x16, centroid_b_y_f32x16);
+        __m512 pb_z_f32x16 = _mm512_sub_ps(b_z_f32x16, centroid_b_z_f32x16);
+
+        __m512 ra_x_f32x16 = _mm512_fmadd_ps(scaled_rotation_x_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_x_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_x_x_f32x16, pa_x_f32x16)));
+        __m512 ra_y_f32x16 = _mm512_fmadd_ps(scaled_rotation_y_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_y_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_y_x_f32x16, pa_x_f32x16)));
+        __m512 ra_z_f32x16 = _mm512_fmadd_ps(scaled_rotation_z_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_z_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_z_x_f32x16, pa_x_f32x16)));
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(ra_x_f32x16, pb_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(ra_y_f32x16, pb_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(ra_z_f32x16, pb_z_f32x16);
+
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_f32x16);
+    }
 
-    for (; index < n; ++index) {
-        nk_f64_t a_x = a[index * 3 + 0], a_y = a[index * 3 + 1], a_z = a[index * 3 + 2];
-        nk_f64_t b_x = b[index * 3 + 0], b_y = b[index * 3 + 1], b_z = b[index * 3 + 2];
-        sum_a_x += a_x, sum_a_y += a_y, sum_a_z += a_z;
-        sum_b_x += b_x, sum_b_y += b_y, sum_b_z += b_z;
-        covariance_00 += a_x * b_x, covariance_01 += a_x * b_y, covariance_02 += a_x * b_z;
-        covariance_10 += a_y * b_x, covariance_11 += a_y * b_y, covariance_12 += a_y * b_z;
-        covariance_20 += a_z * b_x, covariance_21 += a_z * b_y, covariance_22 += a_z * b_z;
+    return _mm512_reduce_add_ps(sum_squared_f32x16);
+}
+
+/*  Compute sum of squared distances for bf16 data after applying rotation (and optional scale).
+ *  Loads bf16, converts to f32 for computation. Rotation matrix, scale, and centroids are f32.
+ */
+NK_INTERNAL nk_f32_t nk_transformed_ssd_bf16_skylake_(nk_bf16_t const *a, nk_bf16_t const *b, nk_size_t n,
+                                                      nk_f32_t const *r, nk_f32_t scale, nk_f32_t centroid_a_x,
+                                                      nk_f32_t centroid_a_y, nk_f32_t centroid_a_z,
+                                                      nk_f32_t centroid_b_x, nk_f32_t centroid_b_y,
+                                                      nk_f32_t centroid_b_z) {
+    __m512 scaled_rotation_x_x_f32x16 = _mm512_set1_ps(scale * r[0]);
+    __m512 scaled_rotation_x_y_f32x16 = _mm512_set1_ps(scale * r[1]);
+    __m512 scaled_rotation_x_z_f32x16 = _mm512_set1_ps(scale * r[2]);
+    __m512 scaled_rotation_y_x_f32x16 = _mm512_set1_ps(scale * r[3]);
+    __m512 scaled_rotation_y_y_f32x16 = _mm512_set1_ps(scale * r[4]);
+    __m512 scaled_rotation_y_z_f32x16 = _mm512_set1_ps(scale * r[5]);
+    __m512 scaled_rotation_z_x_f32x16 = _mm512_set1_ps(scale * r[6]);
+    __m512 scaled_rotation_z_y_f32x16 = _mm512_set1_ps(scale * r[7]);
+    __m512 scaled_rotation_z_z_f32x16 = _mm512_set1_ps(scale * r[8]);
+
+    __m512 centroid_a_x_f32x16 = _mm512_set1_ps(centroid_a_x);
+    __m512 centroid_a_y_f32x16 = _mm512_set1_ps(centroid_a_y);
+    __m512 centroid_a_z_f32x16 = _mm512_set1_ps(centroid_a_z);
+    __m512 centroid_b_x_f32x16 = _mm512_set1_ps(centroid_b_x);
+    __m512 centroid_b_y_f32x16 = _mm512_set1_ps(centroid_b_y);
+    __m512 centroid_b_z_f32x16 = _mm512_set1_ps(centroid_b_z);
+
+    __m512 sum_squared_f32x16 = _mm512_setzero_ps();
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+    nk_size_t j = 0;
+
+    for (; j + 16 <= n; j += 16) {
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(a + j * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(b + j * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        __m512 pa_x_f32x16 = _mm512_sub_ps(a_x_f32x16, centroid_a_x_f32x16);
+        __m512 pa_y_f32x16 = _mm512_sub_ps(a_y_f32x16, centroid_a_y_f32x16);
+        __m512 pa_z_f32x16 = _mm512_sub_ps(a_z_f32x16, centroid_a_z_f32x16);
+        __m512 pb_x_f32x16 = _mm512_sub_ps(b_x_f32x16, centroid_b_x_f32x16);
+        __m512 pb_y_f32x16 = _mm512_sub_ps(b_y_f32x16, centroid_b_y_f32x16);
+        __m512 pb_z_f32x16 = _mm512_sub_ps(b_z_f32x16, centroid_b_z_f32x16);
+
+        __m512 ra_x_f32x16 = _mm512_fmadd_ps(scaled_rotation_x_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_x_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_x_x_f32x16, pa_x_f32x16)));
+        __m512 ra_y_f32x16 = _mm512_fmadd_ps(scaled_rotation_y_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_y_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_y_x_f32x16, pa_x_f32x16)));
+        __m512 ra_z_f32x16 = _mm512_fmadd_ps(scaled_rotation_z_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_z_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_z_x_f32x16, pa_x_f32x16)));
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(ra_x_f32x16, pb_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(ra_y_f32x16, pb_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(ra_z_f32x16, pb_z_f32x16);
+
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_f32x16);
+    }
+
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (j < n) {
+        nk_size_t tail = n - j;
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(a + j * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(b + j * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        __m512 pa_x_f32x16 = _mm512_sub_ps(a_x_f32x16, centroid_a_x_f32x16);
+        __m512 pa_y_f32x16 = _mm512_sub_ps(a_y_f32x16, centroid_a_y_f32x16);
+        __m512 pa_z_f32x16 = _mm512_sub_ps(a_z_f32x16, centroid_a_z_f32x16);
+        __m512 pb_x_f32x16 = _mm512_sub_ps(b_x_f32x16, centroid_b_x_f32x16);
+        __m512 pb_y_f32x16 = _mm512_sub_ps(b_y_f32x16, centroid_b_y_f32x16);
+        __m512 pb_z_f32x16 = _mm512_sub_ps(b_z_f32x16, centroid_b_z_f32x16);
+
+        __m512 ra_x_f32x16 = _mm512_fmadd_ps(scaled_rotation_x_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_x_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_x_x_f32x16, pa_x_f32x16)));
+        __m512 ra_y_f32x16 = _mm512_fmadd_ps(scaled_rotation_y_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_y_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_y_x_f32x16, pa_x_f32x16)));
+        __m512 ra_z_f32x16 = _mm512_fmadd_ps(scaled_rotation_z_z_f32x16, pa_z_f32x16,
+                                             _mm512_fmadd_ps(scaled_rotation_z_y_f32x16, pa_y_f32x16,
+                                                             _mm512_mul_ps(scaled_rotation_z_x_f32x16, pa_x_f32x16)));
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(ra_x_f32x16, pb_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(ra_y_f32x16, pb_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(ra_z_f32x16, pb_z_f32x16);
+
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_f32x16);
+        sum_squared_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_f32x16);
     }
 
-    nk_f64_t inv_n = 1.0 / (nk_f64_t)n, n_f64 = (nk_f64_t)n;
-    *centroid_a_x = sum_a_x * inv_n, *centroid_a_y = sum_a_y * inv_n, *centroid_a_z = sum_a_z * inv_n;
-    *centroid_b_x = sum_b_x * inv_n, *centroid_b_y = sum_b_y * inv_n, *centroid_b_z = sum_b_z * inv_n;
-    cross_covariance_f64[0] = covariance_00 - n_f64 * (*centroid_a_x) * (*centroid_b_x),
-    cross_covariance_f64[1] = covariance_01 - n_f64 * (*centroid_a_x) * (*centroid_b_y),
-    cross_covariance_f64[2] = covariance_02 - n_f64 * (*centroid_a_x) * (*centroid_b_z);
-    cross_covariance_f64[3] = covariance_10 - n_f64 * (*centroid_a_y) * (*centroid_b_x),
-    cross_covariance_f64[4] = covariance_11 - n_f64 * (*centroid_a_y) * (*centroid_b_y),
-    cross_covariance_f64[5] = covariance_12 - n_f64 * (*centroid_a_y) * (*centroid_b_z);
-    cross_covariance_f64[6] = covariance_20 - n_f64 * (*centroid_a_z) * (*centroid_b_x),
-    cross_covariance_f64[7] = covariance_21 - n_f64 * (*centroid_a_z) * (*centroid_b_y),
-    cross_covariance_f64[8] = covariance_22 - n_f64 * (*centroid_a_z) * (*centroid_b_z);
+    return _mm512_reduce_add_ps(sum_squared_f32x16);
 }
 
 NK_PUBLIC void nk_rmsd_f32_skylake(nk_f32_t const *a, nk_f32_t const *b, nk_size_t n, nk_f32_t *a_centroid,
                                    nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f64_t *result) {
-    nk_f64_t identity[9] = {1, 0, 0, 0, 1, 0, 0, 0, 1};
-    nk_f64_t centroid_a_x, centroid_a_y, centroid_a_z, centroid_b_x, centroid_b_y, centroid_b_z;
-    nk_f64_t cross_covariance_f64[9];
     if (rotation)
         rotation[0] = 1, rotation[1] = 0, rotation[2] = 0, rotation[3] = 0, rotation[4] = 1, rotation[5] = 0,
         rotation[6] = 0, rotation[7] = 0, rotation[8] = 1;
     if (scale) *scale = 1.0f;
-    nk_centroid_and_cross_covariance_f32_skylake_(a, b, n, &centroid_a_x, &centroid_a_y, &centroid_a_z, &centroid_b_x,
-                                                  &centroid_b_y, &centroid_b_z, cross_covariance_f64);
+
+    // Fused single-pass: centroids + squared differences in f64, using the identity:
+    //   RMSD = √(E[(a-b)²] - (ā - b̄)²)
+    __m512d const zeros_f64x8 = _mm512_setzero_pd();
+    __m512d sum_a_x_f64x8 = zeros_f64x8, sum_a_y_f64x8 = zeros_f64x8, sum_a_z_f64x8 = zeros_f64x8;
+    __m512d sum_b_x_f64x8 = zeros_f64x8, sum_b_y_f64x8 = zeros_f64x8, sum_b_z_f64x8 = zeros_f64x8;
+    __m512d sum_squared_x_f64x8 = zeros_f64x8, sum_squared_y_f64x8 = zeros_f64x8, sum_squared_z_f64x8 = zeros_f64x8;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+    nk_size_t i = 0;
+
+    // Main loop with 2x unrolling (32 points per iteration)
+    for (; i + 32 <= n; i += 32) {
+        // Iteration 0: points i..i+15
+        nk_deinterleave_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        __m512d delta_x_low_f64x8 = _mm512_sub_pd(a_x_low_f64x8, b_x_low_f64x8);
+        __m512d delta_x_high_f64x8 = _mm512_sub_pd(a_x_high_f64x8, b_x_high_f64x8);
+        __m512d delta_y_low_f64x8 = _mm512_sub_pd(a_y_low_f64x8, b_y_low_f64x8);
+        __m512d delta_y_high_f64x8 = _mm512_sub_pd(a_y_high_f64x8, b_y_high_f64x8);
+        __m512d delta_z_low_f64x8 = _mm512_sub_pd(a_z_low_f64x8, b_z_low_f64x8);
+        __m512d delta_z_high_f64x8 = _mm512_sub_pd(a_z_high_f64x8, b_z_high_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_low_f64x8, delta_x_low_f64x8, sum_squared_x_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_high_f64x8, delta_x_high_f64x8, sum_squared_x_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_low_f64x8, delta_y_low_f64x8, sum_squared_y_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_high_f64x8, delta_y_high_f64x8, sum_squared_y_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_low_f64x8, delta_z_low_f64x8, sum_squared_z_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_high_f64x8, delta_z_high_f64x8, sum_squared_z_f64x8);
+
+        // Iteration 1: points i+16..i+31
+        nk_deinterleave_f32x16_skylake_(a + (i + 16) * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f32x16_skylake_(b + (i + 16) * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+        a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        delta_x_low_f64x8 = _mm512_sub_pd(a_x_low_f64x8, b_x_low_f64x8);
+        delta_x_high_f64x8 = _mm512_sub_pd(a_x_high_f64x8, b_x_high_f64x8);
+        delta_y_low_f64x8 = _mm512_sub_pd(a_y_low_f64x8, b_y_low_f64x8);
+        delta_y_high_f64x8 = _mm512_sub_pd(a_y_high_f64x8, b_y_high_f64x8);
+        delta_z_low_f64x8 = _mm512_sub_pd(a_z_low_f64x8, b_z_low_f64x8);
+        delta_z_high_f64x8 = _mm512_sub_pd(a_z_high_f64x8, b_z_high_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_low_f64x8, delta_x_low_f64x8, sum_squared_x_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_high_f64x8, delta_x_high_f64x8, sum_squared_x_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_low_f64x8, delta_y_low_f64x8, sum_squared_y_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_high_f64x8, delta_y_high_f64x8, sum_squared_y_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_low_f64x8, delta_z_low_f64x8, sum_squared_z_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_high_f64x8, delta_z_high_f64x8, sum_squared_z_f64x8);
+    }
+
+    // Handle 16-point remainder
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        __m512d delta_x_low_f64x8 = _mm512_sub_pd(a_x_low_f64x8, b_x_low_f64x8);
+        __m512d delta_x_high_f64x8 = _mm512_sub_pd(a_x_high_f64x8, b_x_high_f64x8);
+        __m512d delta_y_low_f64x8 = _mm512_sub_pd(a_y_low_f64x8, b_y_low_f64x8);
+        __m512d delta_y_high_f64x8 = _mm512_sub_pd(a_y_high_f64x8, b_y_high_f64x8);
+        __m512d delta_z_low_f64x8 = _mm512_sub_pd(a_z_low_f64x8, b_z_low_f64x8);
+        __m512d delta_z_high_f64x8 = _mm512_sub_pd(a_z_high_f64x8, b_z_high_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_low_f64x8, delta_x_low_f64x8, sum_squared_x_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_high_f64x8, delta_x_high_f64x8, sum_squared_x_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_low_f64x8, delta_y_low_f64x8, sum_squared_y_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_high_f64x8, delta_y_high_f64x8, sum_squared_y_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_low_f64x8, delta_z_low_f64x8, sum_squared_z_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_high_f64x8, delta_z_high_f64x8, sum_squared_z_f64x8);
+    }
+
+    // Tail: use masked gather for remaining < 16 points
+    if (i < n) {
+        nk_size_t tail = n - i;
+        __mmask16 mask = (__mmask16)_bzhi_u32(0xFFFF, tail);
+        __m512i const gather_idx_i32x16 = _mm512_setr_epi32(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45);
+        __m512 zeros_f32x16 = _mm512_setzero_ps();
+        nk_f32_t const *a_tail = a + i * 3;
+        nk_f32_t const *b_tail = b + i * 3;
+
+        a_x_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 0, 4);
+        a_y_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 1, 4);
+        a_z_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 2, 4);
+        b_x_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 0, 4);
+        b_y_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 1, 4);
+        b_z_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 2, 4);
+
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        __m512d delta_x_low_f64x8 = _mm512_sub_pd(a_x_low_f64x8, b_x_low_f64x8);
+        __m512d delta_x_high_f64x8 = _mm512_sub_pd(a_x_high_f64x8, b_x_high_f64x8);
+        __m512d delta_y_low_f64x8 = _mm512_sub_pd(a_y_low_f64x8, b_y_low_f64x8);
+        __m512d delta_y_high_f64x8 = _mm512_sub_pd(a_y_high_f64x8, b_y_high_f64x8);
+        __m512d delta_z_low_f64x8 = _mm512_sub_pd(a_z_low_f64x8, b_z_low_f64x8);
+        __m512d delta_z_high_f64x8 = _mm512_sub_pd(a_z_high_f64x8, b_z_high_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_low_f64x8, delta_x_low_f64x8, sum_squared_x_f64x8);
+        sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_high_f64x8, delta_x_high_f64x8, sum_squared_x_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_low_f64x8, delta_y_low_f64x8, sum_squared_y_f64x8);
+        sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_high_f64x8, delta_y_high_f64x8, sum_squared_y_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_low_f64x8, delta_z_low_f64x8, sum_squared_z_f64x8);
+        sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_high_f64x8, delta_z_high_f64x8, sum_squared_z_f64x8);
+    }
+
+    // Reduce and compute centroids
+    nk_f64_t inv_n = 1.0 / (nk_f64_t)n;
+    nk_f64_t total_ax = _mm512_reduce_add_pd(sum_a_x_f64x8);
+    nk_f64_t total_ay = _mm512_reduce_add_pd(sum_a_y_f64x8);
+    nk_f64_t total_az = _mm512_reduce_add_pd(sum_a_z_f64x8);
+    nk_f64_t total_bx = _mm512_reduce_add_pd(sum_b_x_f64x8);
+    nk_f64_t total_by = _mm512_reduce_add_pd(sum_b_y_f64x8);
+    nk_f64_t total_bz = _mm512_reduce_add_pd(sum_b_z_f64x8);
+    nk_f64_t total_sq_x = _mm512_reduce_add_pd(sum_squared_x_f64x8);
+    nk_f64_t total_sq_y = _mm512_reduce_add_pd(sum_squared_y_f64x8);
+    nk_f64_t total_sq_z = _mm512_reduce_add_pd(sum_squared_z_f64x8);
+
+    nk_f64_t centroid_a_x = total_ax * inv_n, centroid_a_y = total_ay * inv_n, centroid_a_z = total_az * inv_n;
+    nk_f64_t centroid_b_x = total_bx * inv_n, centroid_b_y = total_by * inv_n, centroid_b_z = total_bz * inv_n;
     if (a_centroid)
         a_centroid[0] = (nk_f32_t)centroid_a_x, a_centroid[1] = (nk_f32_t)centroid_a_y,
         a_centroid[2] = (nk_f32_t)centroid_a_z;
     if (b_centroid)
         b_centroid[0] = (nk_f32_t)centroid_b_x, b_centroid[1] = (nk_f32_t)centroid_b_y,
         b_centroid[2] = (nk_f32_t)centroid_b_z;
-    *result = nk_f64_sqrt_haswell(nk_transformed_ssd_f32_skylake_(a, b, n, identity, 1.0, centroid_a_x, centroid_a_y,
-                                                                  centroid_a_z, centroid_b_x, centroid_b_y,
-                                                                  centroid_b_z) /
-                                  (nk_f64_t)n);
+
+    nk_f64_t mean_diff_x = centroid_a_x - centroid_b_x, mean_diff_y = centroid_a_y - centroid_b_y,
+             mean_diff_z = centroid_a_z - centroid_b_z;
+    nk_f64_t sum_squared = total_sq_x + total_sq_y + total_sq_z;
+    nk_f64_t mean_diff_sq = mean_diff_x * mean_diff_x + mean_diff_y * mean_diff_y + mean_diff_z * mean_diff_z;
+    *result = nk_f64_sqrt_haswell(sum_squared * inv_n - mean_diff_sq);
 }
 
 NK_PUBLIC void nk_kabsch_f32_skylake(nk_f32_t const *a, nk_f32_t const *b, nk_size_t n, nk_f32_t *a_centroid,
                                      nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f64_t *result) {
-    nk_f64_t centroid_a_x, centroid_a_y, centroid_a_z, centroid_b_x, centroid_b_y, centroid_b_z;
-    nk_f64_t cross_covariance_f64[9];
-    nk_centroid_and_cross_covariance_f32_skylake_(a, b, n, &centroid_a_x, &centroid_a_y, &centroid_a_z, &centroid_b_x,
-                                                  &centroid_b_y, &centroid_b_z, cross_covariance_f64);
+    // Fused single-pass: centroids + covariance in f64
+    __m512d const zeros_f64x8 = _mm512_setzero_pd();
+    __m512d sum_a_x_f64x8 = zeros_f64x8, sum_a_y_f64x8 = zeros_f64x8, sum_a_z_f64x8 = zeros_f64x8;
+    __m512d sum_b_x_f64x8 = zeros_f64x8, sum_b_y_f64x8 = zeros_f64x8, sum_b_z_f64x8 = zeros_f64x8;
+    __m512d cov_xx_f64x8 = zeros_f64x8, cov_xy_f64x8 = zeros_f64x8, cov_xz_f64x8 = zeros_f64x8;
+    __m512d cov_yx_f64x8 = zeros_f64x8, cov_yy_f64x8 = zeros_f64x8, cov_yz_f64x8 = zeros_f64x8;
+    __m512d cov_zx_f64x8 = zeros_f64x8, cov_zy_f64x8 = zeros_f64x8, cov_zz_f64x8 = zeros_f64x8;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+    nk_size_t i = 0;
+
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        cov_xx_f64x8 = _mm512_add_pd(cov_xx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_x_high_f64x8)));
+        cov_xy_f64x8 = _mm512_add_pd(cov_xy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_y_high_f64x8)));
+        cov_xz_f64x8 = _mm512_add_pd(cov_xz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_z_high_f64x8)));
+        cov_yx_f64x8 = _mm512_add_pd(cov_yx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_x_high_f64x8)));
+        cov_yy_f64x8 = _mm512_add_pd(cov_yy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_y_high_f64x8)));
+        cov_yz_f64x8 = _mm512_add_pd(cov_yz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_z_high_f64x8)));
+        cov_zx_f64x8 = _mm512_add_pd(cov_zx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_x_high_f64x8)));
+        cov_zy_f64x8 = _mm512_add_pd(cov_zy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_y_high_f64x8)));
+        cov_zz_f64x8 = _mm512_add_pd(cov_zz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_z_high_f64x8)));
+    }
+
+    // Tail: use masked gather for remaining < 16 points
+    if (i < n) {
+        nk_size_t tail = n - i;
+        __mmask16 mask = (__mmask16)_bzhi_u32(0xFFFF, tail);
+        __m512i const gather_idx_i32x16 = _mm512_setr_epi32(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45);
+        __m512 zeros_f32x16 = _mm512_setzero_ps();
+        nk_f32_t const *a_tail = a + i * 3;
+        nk_f32_t const *b_tail = b + i * 3;
+
+        a_x_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 0, 4);
+        a_y_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 1, 4);
+        a_z_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 2, 4);
+        b_x_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 0, 4);
+        b_y_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 1, 4);
+        b_z_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 2, 4);
+
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        cov_xx_f64x8 = _mm512_add_pd(cov_xx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_x_high_f64x8)));
+        cov_xy_f64x8 = _mm512_add_pd(cov_xy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_y_high_f64x8)));
+        cov_xz_f64x8 = _mm512_add_pd(cov_xz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_z_high_f64x8)));
+        cov_yx_f64x8 = _mm512_add_pd(cov_yx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_x_high_f64x8)));
+        cov_yy_f64x8 = _mm512_add_pd(cov_yy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_y_high_f64x8)));
+        cov_yz_f64x8 = _mm512_add_pd(cov_yz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_z_high_f64x8)));
+        cov_zx_f64x8 = _mm512_add_pd(cov_zx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_x_high_f64x8)));
+        cov_zy_f64x8 = _mm512_add_pd(cov_zy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_y_high_f64x8)));
+        cov_zz_f64x8 = _mm512_add_pd(cov_zz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_z_high_f64x8)));
+    }
+
+    nk_f64_t sum_a_x = _mm512_reduce_add_pd(sum_a_x_f64x8), sum_a_y = _mm512_reduce_add_pd(sum_a_y_f64x8),
+             sum_a_z = _mm512_reduce_add_pd(sum_a_z_f64x8);
+    nk_f64_t sum_b_x = _mm512_reduce_add_pd(sum_b_x_f64x8), sum_b_y = _mm512_reduce_add_pd(sum_b_y_f64x8),
+             sum_b_z = _mm512_reduce_add_pd(sum_b_z_f64x8);
+    nk_f64_t covariance_x_x = _mm512_reduce_add_pd(cov_xx_f64x8), covariance_x_y = _mm512_reduce_add_pd(cov_xy_f64x8),
+             covariance_x_z = _mm512_reduce_add_pd(cov_xz_f64x8);
+    nk_f64_t covariance_y_x = _mm512_reduce_add_pd(cov_yx_f64x8), covariance_y_y = _mm512_reduce_add_pd(cov_yy_f64x8),
+             covariance_y_z = _mm512_reduce_add_pd(cov_yz_f64x8);
+    nk_f64_t covariance_z_x = _mm512_reduce_add_pd(cov_zx_f64x8), covariance_z_y = _mm512_reduce_add_pd(cov_zy_f64x8),
+             covariance_z_z = _mm512_reduce_add_pd(cov_zz_f64x8);
+
+    nk_f64_t inv_n = 1.0 / (nk_f64_t)n;
+    nk_f64_t centroid_a_x = sum_a_x * inv_n, centroid_a_y = sum_a_y * inv_n, centroid_a_z = sum_a_z * inv_n;
+    nk_f64_t centroid_b_x = sum_b_x * inv_n, centroid_b_y = sum_b_y * inv_n, centroid_b_z = sum_b_z * inv_n;
     if (a_centroid)
         a_centroid[0] = (nk_f32_t)centroid_a_x, a_centroid[1] = (nk_f32_t)centroid_a_y,
         a_centroid[2] = (nk_f32_t)centroid_a_z;
@@ -485,51 +978,40 @@ NK_PUBLIC void nk_kabsch_f32_skylake(nk_f32_t const *a, nk_f32_t const *b, nk_si
         b_centroid[2] = (nk_f32_t)centroid_b_z;
     if (scale) *scale = 1.0f;
 
-    nk_f64_t svd_u[9], svd_s[9], svd_v[9], r[9];
-    nk_svd3x3_f64_(cross_covariance_f64, svd_u, svd_s, svd_v);
-    r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
-    r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
-    r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
-    r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
-    r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
-    r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
-    r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
-    r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
-    r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+    nk_f64_t n_f64 = (nk_f64_t)n;
+    nk_f64_t cross_covariance[9];
+    cross_covariance[0] = covariance_x_x - n_f64 * centroid_a_x * centroid_b_x;
+    cross_covariance[1] = covariance_x_y - n_f64 * centroid_a_x * centroid_b_y;
+    cross_covariance[2] = covariance_x_z - n_f64 * centroid_a_x * centroid_b_z;
+    cross_covariance[3] = covariance_y_x - n_f64 * centroid_a_y * centroid_b_x;
+    cross_covariance[4] = covariance_y_y - n_f64 * centroid_a_y * centroid_b_y;
+    cross_covariance[5] = covariance_y_z - n_f64 * centroid_a_y * centroid_b_z;
+    cross_covariance[6] = covariance_z_x - n_f64 * centroid_a_z * centroid_b_x;
+    cross_covariance[7] = covariance_z_y - n_f64 * centroid_a_z * centroid_b_y;
+    cross_covariance[8] = covariance_z_z - n_f64 * centroid_a_z * centroid_b_z;
+
+    nk_f64_t svd_u[9], svd_s[9], svd_v[9];
+    nk_svd3x3_f64_(cross_covariance, svd_u, svd_s, svd_v);
+    nk_f64_t r[9];
+    nk_rotation_from_svd_f64_skylake_(svd_u, svd_v, r);
     if (nk_det3x3_f64_(r) < 0) {
         svd_v[2] = -svd_v[2], svd_v[5] = -svd_v[5], svd_v[8] = -svd_v[8];
-        r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
-        r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
-        r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
-        r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
-        r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
-        r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
-        r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
-        r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
-        r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+        nk_rotation_from_svd_f64_skylake_(svd_u, svd_v, r);
     }
     if (rotation)
-        for (int index = 0; index != 9; ++index) rotation[index] = (nk_f32_t)r[index];
+        for (int j = 0; j < 9; ++j) rotation[j] = (nk_f32_t)r[j];
     *result = nk_f64_sqrt_haswell(nk_transformed_ssd_f32_skylake_(a, b, n, r, 1.0, centroid_a_x, centroid_a_y,
                                                                   centroid_a_z, centroid_b_x, centroid_b_y,
                                                                   centroid_b_z) /
-                                  (nk_f64_t)n);
+                                  n_f64);
 }
 
 NK_PUBLIC void nk_rmsd_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_size_t n, nk_f64_t *a_centroid,
                                    nk_f64_t *b_centroid, nk_f64_t *rotation, nk_f64_t *scale, nk_f64_t *result) {
     // RMSD uses identity rotation and scale=1.0.
-    if (rotation) {
-        rotation[0] = 1;
-        rotation[1] = 0;
-        rotation[2] = 0;
-        rotation[3] = 0;
-        rotation[4] = 1;
-        rotation[5] = 0;
-        rotation[6] = 0;
-        rotation[7] = 0;
-        rotation[8] = 1;
-    }
+    if (rotation)
+        rotation[0] = 1, rotation[1] = 0, rotation[2] = 0, rotation[3] = 0, rotation[4] = 1, rotation[5] = 0,
+        rotation[6] = 0, rotation[7] = 0, rotation[8] = 1;
     if (scale) *scale = 1.0;
     // Optimized fused single-pass implementation for f64.
     // Computes centroids and squared differences in one pass using the identity:
@@ -633,6 +1115,7 @@ NK_PUBLIC void nk_rmsd_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_size
         sum_squared_x_f64x8 = _mm512_fmadd_pd(delta_x_f64x8, delta_x_f64x8, sum_squared_x_f64x8);
         sum_squared_y_f64x8 = _mm512_fmadd_pd(delta_y_f64x8, delta_y_f64x8, sum_squared_y_f64x8);
         sum_squared_z_f64x8 = _mm512_fmadd_pd(delta_z_f64x8, delta_z_f64x8, sum_squared_z_f64x8);
+        i = n;
     }
 
     // Reduce and compute centroids.
@@ -759,6 +1242,7 @@ NK_PUBLIC void nk_kabsch_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_si
         cov_zx_f64x8 = _mm512_fmadd_pd(a_z_f64x8, b_x_f64x8, cov_zx_f64x8),
         cov_zy_f64x8 = _mm512_fmadd_pd(a_z_f64x8, b_y_f64x8, cov_zy_f64x8),
         cov_zz_f64x8 = _mm512_fmadd_pd(a_z_f64x8, b_z_f64x8, cov_zz_f64x8);
+        i = n;
     }
 
     // Reduce centroids and covariance.
@@ -840,9 +1324,8 @@ NK_PUBLIC void nk_kabsch_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_si
     }
 
     // Output rotation matrix and scale=1.0.
-    if (rotation) {
+    if (rotation)
         for (int j = 0; j < 9; ++j) rotation[j] = (nk_f64_t)r[j];
-    }
     if (scale) *scale = 1.0;
 
     // Compute RMSD after optimal rotation
@@ -851,51 +1334,153 @@ NK_PUBLIC void nk_kabsch_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_si
     *result = nk_f64_sqrt_haswell(sum_squared * inv_n);
 }
 
-NK_INTERNAL void nk_centroid_and_cross_covariance_and_variance_f32_skylake_( //
-    nk_f32_t const *a, nk_f32_t const *b, nk_size_t n,                       //
-    nk_f64_t *centroid_a_x, nk_f64_t *centroid_a_y, nk_f64_t *centroid_a_z, nk_f64_t *centroid_b_x,
-    nk_f64_t *centroid_b_y, nk_f64_t *centroid_b_z, nk_f64_t cross_covariance_f64[9], nk_f64_t *variance_a) {
-    nk_centroid_and_cross_covariance_f32_skylake_(a, b, n, centroid_a_x, centroid_a_y, centroid_a_z, centroid_b_x,
-                                                  centroid_b_y, centroid_b_z, cross_covariance_f64);
-    __m512d variance_a_f64x8 = _mm512_setzero_pd();
-    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16;
-    nk_size_t index = 0;
+NK_PUBLIC void nk_umeyama_f32_skylake(nk_f32_t const *a, nk_f32_t const *b, nk_size_t n, nk_f32_t *a_centroid,
+                                      nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f64_t *result) {
+    // Fused single-pass: centroids + covariance + variance of A, all in f64
+    __m512d const zeros_f64x8 = _mm512_setzero_pd();
+    __m512d sum_a_x_f64x8 = zeros_f64x8, sum_a_y_f64x8 = zeros_f64x8, sum_a_z_f64x8 = zeros_f64x8;
+    __m512d sum_b_x_f64x8 = zeros_f64x8, sum_b_y_f64x8 = zeros_f64x8, sum_b_z_f64x8 = zeros_f64x8;
+    __m512d cov_xx_f64x8 = zeros_f64x8, cov_xy_f64x8 = zeros_f64x8, cov_xz_f64x8 = zeros_f64x8;
+    __m512d cov_yx_f64x8 = zeros_f64x8, cov_yy_f64x8 = zeros_f64x8, cov_yz_f64x8 = zeros_f64x8;
+    __m512d cov_zx_f64x8 = zeros_f64x8, cov_zy_f64x8 = zeros_f64x8, cov_zz_f64x8 = zeros_f64x8;
+    __m512d variance_a_f64x8 = zeros_f64x8;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+    nk_size_t i = 0;
 
-    for (; index + 16 <= n; index += 16) {
-        nk_deinterleave_f32x16_skylake_(a + index * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
-        __m512d a_x_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
-        __m512d a_x_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
-        __m512d a_y_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
-        __m512d a_y_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
-        __m512d a_z_lower_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
-        __m512d a_z_upper_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
-        __m512d batch_norm_squared_f64x8 = _mm512_add_pd(_mm512_mul_pd(a_x_lower_f64x8, a_x_lower_f64x8),
-                                                         _mm512_mul_pd(a_x_upper_f64x8, a_x_upper_f64x8));
-        batch_norm_squared_f64x8 = _mm512_fmadd_pd(a_y_lower_f64x8, a_y_lower_f64x8, batch_norm_squared_f64x8);
-        batch_norm_squared_f64x8 = _mm512_fmadd_pd(a_y_upper_f64x8, a_y_upper_f64x8, batch_norm_squared_f64x8);
-        batch_norm_squared_f64x8 = _mm512_fmadd_pd(a_z_lower_f64x8, a_z_lower_f64x8, batch_norm_squared_f64x8);
-        batch_norm_squared_f64x8 = _mm512_fmadd_pd(a_z_upper_f64x8, a_z_upper_f64x8, batch_norm_squared_f64x8);
-        variance_a_f64x8 = _mm512_add_pd(variance_a_f64x8, batch_norm_squared_f64x8);
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        cov_xx_f64x8 = _mm512_add_pd(cov_xx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_x_high_f64x8)));
+        cov_xy_f64x8 = _mm512_add_pd(cov_xy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_y_high_f64x8)));
+        cov_xz_f64x8 = _mm512_add_pd(cov_xz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_z_high_f64x8)));
+        cov_yx_f64x8 = _mm512_add_pd(cov_yx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_x_high_f64x8)));
+        cov_yy_f64x8 = _mm512_add_pd(cov_yy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_y_high_f64x8)));
+        cov_yz_f64x8 = _mm512_add_pd(cov_yz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_z_high_f64x8)));
+        cov_zx_f64x8 = _mm512_add_pd(cov_zx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_x_high_f64x8)));
+        cov_zy_f64x8 = _mm512_add_pd(cov_zy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_y_high_f64x8)));
+        cov_zz_f64x8 = _mm512_add_pd(cov_zz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_z_high_f64x8)));
+
+        variance_a_f64x8 = _mm512_add_pd(
+            variance_a_f64x8,
+            _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, a_x_low_f64x8), _mm512_mul_pd(a_x_high_f64x8, a_x_high_f64x8)));
+        variance_a_f64x8 = _mm512_add_pd(
+            variance_a_f64x8,
+            _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, a_y_low_f64x8), _mm512_mul_pd(a_y_high_f64x8, a_y_high_f64x8)));
+        variance_a_f64x8 = _mm512_add_pd(
+            variance_a_f64x8,
+            _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, a_z_low_f64x8), _mm512_mul_pd(a_z_high_f64x8, a_z_high_f64x8)));
     }
 
-    nk_f64_t variance_sum = _mm512_reduce_add_pd(variance_a_f64x8);
-    for (; index < n; ++index) {
-        nk_f64_t a_x = a[index * 3 + 0], a_y = a[index * 3 + 1], a_z = a[index * 3 + 2];
-        variance_sum += a_x * a_x + a_y * a_y + a_z * a_z;
+    // Tail: use masked gather for remaining < 16 points
+    if (i < n) {
+        nk_size_t tail = n - i;
+        __mmask16 mask = (__mmask16)_bzhi_u32(0xFFFF, tail);
+        __m512i const gather_idx_i32x16 = _mm512_setr_epi32(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45);
+        __m512 zeros_f32x16 = _mm512_setzero_ps();
+        nk_f32_t const *a_tail = a + i * 3;
+        nk_f32_t const *b_tail = b + i * 3;
+
+        a_x_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 0, 4);
+        a_y_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 1, 4);
+        a_z_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, a_tail + 2, 4);
+        b_x_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 0, 4);
+        b_y_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 1, 4);
+        b_z_f32x16 = _mm512_mask_i32gather_ps(zeros_f32x16, mask, gather_idx_i32x16, b_tail + 2, 4);
+
+        __m512d a_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_x_f32x16));
+        __m512d a_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_x_f32x16, 1));
+        __m512d a_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_y_f32x16));
+        __m512d a_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_y_f32x16, 1));
+        __m512d a_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(a_z_f32x16));
+        __m512d a_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(a_z_f32x16, 1));
+        __m512d b_x_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_x_f32x16));
+        __m512d b_x_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_x_f32x16, 1));
+        __m512d b_y_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_y_f32x16));
+        __m512d b_y_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_y_f32x16, 1));
+        __m512d b_z_low_f64x8 = _mm512_cvtps_pd(_mm512_castps512_ps256(b_z_f32x16));
+        __m512d b_z_high_f64x8 = _mm512_cvtps_pd(_mm512_extractf32x8_ps(b_z_f32x16, 1));
+
+        sum_a_x_f64x8 = _mm512_add_pd(sum_a_x_f64x8, _mm512_add_pd(a_x_low_f64x8, a_x_high_f64x8));
+        sum_a_y_f64x8 = _mm512_add_pd(sum_a_y_f64x8, _mm512_add_pd(a_y_low_f64x8, a_y_high_f64x8));
+        sum_a_z_f64x8 = _mm512_add_pd(sum_a_z_f64x8, _mm512_add_pd(a_z_low_f64x8, a_z_high_f64x8));
+        sum_b_x_f64x8 = _mm512_add_pd(sum_b_x_f64x8, _mm512_add_pd(b_x_low_f64x8, b_x_high_f64x8));
+        sum_b_y_f64x8 = _mm512_add_pd(sum_b_y_f64x8, _mm512_add_pd(b_y_low_f64x8, b_y_high_f64x8));
+        sum_b_z_f64x8 = _mm512_add_pd(sum_b_z_f64x8, _mm512_add_pd(b_z_low_f64x8, b_z_high_f64x8));
+
+        cov_xx_f64x8 = _mm512_add_pd(cov_xx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_x_high_f64x8)));
+        cov_xy_f64x8 = _mm512_add_pd(cov_xy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_y_high_f64x8)));
+        cov_xz_f64x8 = _mm512_add_pd(cov_xz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_x_high_f64x8, b_z_high_f64x8)));
+        cov_yx_f64x8 = _mm512_add_pd(cov_yx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_x_high_f64x8)));
+        cov_yy_f64x8 = _mm512_add_pd(cov_yy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_y_high_f64x8)));
+        cov_yz_f64x8 = _mm512_add_pd(cov_yz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_y_high_f64x8, b_z_high_f64x8)));
+        cov_zx_f64x8 = _mm512_add_pd(cov_zx_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_x_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_x_high_f64x8)));
+        cov_zy_f64x8 = _mm512_add_pd(cov_zy_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_y_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_y_high_f64x8)));
+        cov_zz_f64x8 = _mm512_add_pd(cov_zz_f64x8, _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, b_z_low_f64x8),
+                                                                 _mm512_mul_pd(a_z_high_f64x8, b_z_high_f64x8)));
+
+        variance_a_f64x8 = _mm512_add_pd(
+            variance_a_f64x8,
+            _mm512_add_pd(_mm512_mul_pd(a_x_low_f64x8, a_x_low_f64x8), _mm512_mul_pd(a_x_high_f64x8, a_x_high_f64x8)));
+        variance_a_f64x8 = _mm512_add_pd(
+            variance_a_f64x8,
+            _mm512_add_pd(_mm512_mul_pd(a_y_low_f64x8, a_y_low_f64x8), _mm512_mul_pd(a_y_high_f64x8, a_y_high_f64x8)));
+        variance_a_f64x8 = _mm512_add_pd(
+            variance_a_f64x8,
+            _mm512_add_pd(_mm512_mul_pd(a_z_low_f64x8, a_z_low_f64x8), _mm512_mul_pd(a_z_high_f64x8, a_z_high_f64x8)));
     }
 
-    nk_f64_t inv_n = 1.0 / (nk_f64_t)n;
-    *variance_a = variance_sum * inv_n - ((*centroid_a_x) * (*centroid_a_x) + (*centroid_a_y) * (*centroid_a_y) +
-                                          (*centroid_a_z) * (*centroid_a_z));
-}
+    nk_f64_t sum_a_x = _mm512_reduce_add_pd(sum_a_x_f64x8), sum_a_y = _mm512_reduce_add_pd(sum_a_y_f64x8),
+             sum_a_z = _mm512_reduce_add_pd(sum_a_z_f64x8);
+    nk_f64_t sum_b_x = _mm512_reduce_add_pd(sum_b_x_f64x8), sum_b_y = _mm512_reduce_add_pd(sum_b_y_f64x8),
+             sum_b_z = _mm512_reduce_add_pd(sum_b_z_f64x8);
+    nk_f64_t covariance_x_x = _mm512_reduce_add_pd(cov_xx_f64x8), covariance_x_y = _mm512_reduce_add_pd(cov_xy_f64x8),
+             covariance_x_z = _mm512_reduce_add_pd(cov_xz_f64x8);
+    nk_f64_t covariance_y_x = _mm512_reduce_add_pd(cov_yx_f64x8), covariance_y_y = _mm512_reduce_add_pd(cov_yy_f64x8),
+             covariance_y_z = _mm512_reduce_add_pd(cov_yz_f64x8);
+    nk_f64_t covariance_z_x = _mm512_reduce_add_pd(cov_zx_f64x8), covariance_z_y = _mm512_reduce_add_pd(cov_zy_f64x8),
+             covariance_z_z = _mm512_reduce_add_pd(cov_zz_f64x8);
+    nk_f64_t variance_a_sum = _mm512_reduce_add_pd(variance_a_f64x8);
 
-NK_PUBLIC void nk_umeyama_f32_skylake(nk_f32_t const *a, nk_f32_t const *b, nk_size_t n, nk_f32_t *a_centroid,
-                                      nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f64_t *result) {
-    nk_f64_t centroid_a_x, centroid_a_y, centroid_a_z, centroid_b_x, centroid_b_y, centroid_b_z, variance_a;
-    nk_f64_t cross_covariance_f64[9];
-    nk_centroid_and_cross_covariance_and_variance_f32_skylake_(a, b, n, &centroid_a_x, &centroid_a_y, &centroid_a_z,
-                                                               &centroid_b_x, &centroid_b_y, &centroid_b_z,
-                                                               cross_covariance_f64, &variance_a);
+    nk_f64_t inv_n = 1.0 / (nk_f64_t)n;
+    nk_f64_t centroid_a_x = sum_a_x * inv_n, centroid_a_y = sum_a_y * inv_n, centroid_a_z = sum_a_z * inv_n;
+    nk_f64_t centroid_b_x = sum_b_x * inv_n, centroid_b_y = sum_b_y * inv_n, centroid_b_z = sum_b_z * inv_n;
     if (a_centroid)
         a_centroid[0] = (nk_f32_t)centroid_a_x, a_centroid[1] = (nk_f32_t)centroid_a_y,
         a_centroid[2] = (nk_f32_t)centroid_a_z;
@@ -903,41 +1488,49 @@ NK_PUBLIC void nk_umeyama_f32_skylake(nk_f32_t const *a, nk_f32_t const *b, nk_s
         b_centroid[0] = (nk_f32_t)centroid_b_x, b_centroid[1] = (nk_f32_t)centroid_b_y,
         b_centroid[2] = (nk_f32_t)centroid_b_z;
 
-    nk_f64_t svd_u[9], svd_s[9], svd_v[9], r[9];
-    nk_svd3x3_f64_(cross_covariance_f64, svd_u, svd_s, svd_v);
-    r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
-    r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
-    r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
-    r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
-    r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
-    r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
-    r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
-    r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
-    r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+    // Compute centered covariance and variance
+    nk_f64_t variance_a = variance_a_sum * inv_n -
+                          (centroid_a_x * centroid_a_x + centroid_a_y * centroid_a_y + centroid_a_z * centroid_a_z);
 
+    // Compute centered covariance matrix: Hᵢⱼ = Σ(aᵢ×bⱼ) - Σaᵢ × Σbⱼ / n
+    nk_f64_t n_f64 = (nk_f64_t)n;
+    nk_f64_t cross_covariance[9];
+    cross_covariance[0] = covariance_x_x - n_f64 * centroid_a_x * centroid_b_x;
+    cross_covariance[1] = covariance_x_y - n_f64 * centroid_a_x * centroid_b_y;
+    cross_covariance[2] = covariance_x_z - n_f64 * centroid_a_x * centroid_b_z;
+    cross_covariance[3] = covariance_y_x - n_f64 * centroid_a_y * centroid_b_x;
+    cross_covariance[4] = covariance_y_y - n_f64 * centroid_a_y * centroid_b_y;
+    cross_covariance[5] = covariance_y_z - n_f64 * centroid_a_y * centroid_b_z;
+    cross_covariance[6] = covariance_z_x - n_f64 * centroid_a_z * centroid_b_x;
+    cross_covariance[7] = covariance_z_y - n_f64 * centroid_a_z * centroid_b_y;
+    cross_covariance[8] = covariance_z_z - n_f64 * centroid_a_z * centroid_b_z;
+
+    // SVD using f64 for full precision
+    nk_f64_t svd_u[9], svd_s[9], svd_v[9];
+    nk_svd3x3_f64_(cross_covariance, svd_u, svd_s, svd_v);
+
+    nk_f64_t r[9];
+    nk_rotation_from_svd_f64_skylake_(svd_u, svd_v, r);
+
+    // Scale factor: c = trace(D × S) / (n × variance(a))
     nk_f64_t det = nk_det3x3_f64_(r);
-    nk_f64_t trace_signed_singular_values = svd_s[0] + svd_s[4] + (det < 0 ? -svd_s[8] : svd_s[8]);
-    nk_f64_t applied_scale = trace_signed_singular_values / ((nk_f64_t)n * variance_a);
+    nk_f64_t d3 = det < 0 ? -1.0 : 1.0;
+    nk_f64_t trace_ds = nk_sum_three_products_f64_(svd_s[0], 1.0, svd_s[4], 1.0, svd_s[8], d3);
+    nk_f64_t applied_scale = trace_ds / ((nk_f64_t)n * variance_a);
+    if (scale) *scale = (nk_f32_t)applied_scale;
+
+    // Handle reflection
     if (det < 0) {
         svd_v[2] = -svd_v[2], svd_v[5] = -svd_v[5], svd_v[8] = -svd_v[8];
-        r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
-        r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
-        r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
-        r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
-        r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
-        r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
-        r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
-        r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
-        r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+        nk_rotation_from_svd_f64_skylake_(svd_u, svd_v, r);
     }
 
     if (rotation)
-        for (int index = 0; index != 9; ++index) rotation[index] = (nk_f32_t)r[index];
-    if (scale) *scale = (nk_f32_t)applied_scale;
+        for (int j = 0; j < 9; ++j) rotation[j] = (nk_f32_t)r[j];
     *result = nk_f64_sqrt_haswell(nk_transformed_ssd_f32_skylake_(a, b, n, r, applied_scale, centroid_a_x, centroid_a_y,
                                                                   centroid_a_z, centroid_b_x, centroid_b_y,
                                                                   centroid_b_z) /
-                                  (nk_f64_t)n);
+                                  n_f64);
 }
 
 NK_PUBLIC void nk_umeyama_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_size_t n, nk_f64_t *a_centroid,
@@ -1013,6 +1606,7 @@ NK_PUBLIC void nk_umeyama_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_s
         variance_a_f64x8 = _mm512_fmadd_pd(a_x_f64x8, a_x_f64x8, variance_a_f64x8);
         variance_a_f64x8 = _mm512_fmadd_pd(a_y_f64x8, a_y_f64x8, variance_a_f64x8);
         variance_a_f64x8 = _mm512_fmadd_pd(a_z_f64x8, a_z_f64x8, variance_a_f64x8);
+        i = n;
     }
 
     // Reduce centroids, covariance, and variance.
@@ -1100,7 +1694,7 @@ NK_PUBLIC void nk_umeyama_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_s
     nk_f64_t det = nk_det3x3_f64_(r);
     nk_f64_t d3 = det < 0 ? -1.0 : 1.0;
     nk_f64_t trace_ds = nk_sum_three_products_f64_(svd_s[0], 1.0, svd_s[4], 1.0, svd_s[8], d3);
-    nk_f64_t c = trace_ds / (n * variance_a);
+    nk_f64_t c = trace_ds / ((nk_f64_t)n * variance_a);
     if (scale) *scale = c;
 
     // Handle reflection
@@ -1110,9 +1704,8 @@ NK_PUBLIC void nk_umeyama_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_s
     }
 
     // Output rotation matrix.
-    if (rotation) {
+    if (rotation)
         for (int j = 0; j < 9; ++j) rotation[j] = (nk_f64_t)r[j];
-    }
 
     // Compute RMSD with scaling
     nk_f64_t sum_squared = nk_transformed_ssd_f64_skylake_(a, b, n, r, c, centroid_a_x, centroid_a_y, centroid_a_z,
@@ -1120,6 +1713,738 @@ NK_PUBLIC void nk_umeyama_f64_skylake(nk_f64_t const *a, nk_f64_t const *b, nk_s
     *result = nk_f64_sqrt_haswell(sum_squared * inv_n);
 }
 
+NK_PUBLIC void nk_rmsd_f16_skylake(nk_f16_t const *a, nk_f16_t const *b, nk_size_t n, nk_f32_t *a_centroid,
+                                   nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f32_t *result) {
+    if (rotation)
+        rotation[0] = 1, rotation[1] = 0, rotation[2] = 0, rotation[3] = 0, rotation[4] = 1, rotation[5] = 0,
+        rotation[6] = 0, rotation[7] = 0, rotation[8] = 1;
+    if (scale) *scale = 1.0f;
+
+    __m512 const zeros_f32x16 = _mm512_setzero_ps();
+    __m512 sum_a_x_f32x16 = zeros_f32x16, sum_a_y_f32x16 = zeros_f32x16, sum_a_z_f32x16 = zeros_f32x16;
+    __m512 sum_b_x_f32x16 = zeros_f32x16, sum_b_y_f32x16 = zeros_f32x16, sum_b_z_f32x16 = zeros_f32x16;
+    __m512 sum_squared_x_f32x16 = zeros_f32x16, sum_squared_y_f32x16 = zeros_f32x16;
+    __m512 sum_squared_z_f32x16 = zeros_f32x16;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+    nk_size_t i = 0;
+
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_f16x16_to_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16x16_to_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(a_x_f32x16, b_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(a_y_f32x16, b_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(a_z_f32x16, b_z_f32x16);
+
+        sum_squared_x_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_x_f32x16);
+        sum_squared_y_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_y_f32x16);
+        sum_squared_z_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_z_f32x16);
+    }
+
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (i < n) {
+        nk_size_t tail = n - i;
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(a + i * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(b + i * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(a_x_f32x16, b_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(a_y_f32x16, b_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(a_z_f32x16, b_z_f32x16);
+
+        sum_squared_x_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_x_f32x16);
+        sum_squared_y_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_y_f32x16);
+        sum_squared_z_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_z_f32x16);
+    }
+
+    nk_f32_t total_ax = _mm512_reduce_add_ps(sum_a_x_f32x16);
+    nk_f32_t total_ay = _mm512_reduce_add_ps(sum_a_y_f32x16);
+    nk_f32_t total_az = _mm512_reduce_add_ps(sum_a_z_f32x16);
+    nk_f32_t total_bx = _mm512_reduce_add_ps(sum_b_x_f32x16);
+    nk_f32_t total_by = _mm512_reduce_add_ps(sum_b_y_f32x16);
+    nk_f32_t total_bz = _mm512_reduce_add_ps(sum_b_z_f32x16);
+    nk_f32_t total_sq_x = _mm512_reduce_add_ps(sum_squared_x_f32x16);
+    nk_f32_t total_sq_y = _mm512_reduce_add_ps(sum_squared_y_f32x16);
+    nk_f32_t total_sq_z = _mm512_reduce_add_ps(sum_squared_z_f32x16);
+
+    nk_f32_t inv_n = 1.0f / (nk_f32_t)n;
+    nk_f32_t centroid_a_x = total_ax * inv_n;
+    nk_f32_t centroid_a_y = total_ay * inv_n;
+    nk_f32_t centroid_a_z = total_az * inv_n;
+    nk_f32_t centroid_b_x = total_bx * inv_n;
+    nk_f32_t centroid_b_y = total_by * inv_n;
+    nk_f32_t centroid_b_z = total_bz * inv_n;
+
+    if (a_centroid) a_centroid[0] = centroid_a_x, a_centroid[1] = centroid_a_y, a_centroid[2] = centroid_a_z;
+    if (b_centroid) b_centroid[0] = centroid_b_x, b_centroid[1] = centroid_b_y, b_centroid[2] = centroid_b_z;
+
+    nk_f32_t mean_diff_x = centroid_a_x - centroid_b_x;
+    nk_f32_t mean_diff_y = centroid_a_y - centroid_b_y;
+    nk_f32_t mean_diff_z = centroid_a_z - centroid_b_z;
+    nk_f32_t sum_squared = total_sq_x + total_sq_y + total_sq_z;
+    nk_f32_t mean_diff_sq = mean_diff_x * mean_diff_x + mean_diff_y * mean_diff_y + mean_diff_z * mean_diff_z;
+
+    *result = nk_f32_sqrt_haswell(sum_squared * inv_n - mean_diff_sq);
+}
+
+NK_PUBLIC void nk_rmsd_bf16_skylake(nk_bf16_t const *a, nk_bf16_t const *b, nk_size_t n, nk_f32_t *a_centroid,
+                                    nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f32_t *result) {
+    if (rotation)
+        rotation[0] = 1, rotation[1] = 0, rotation[2] = 0, rotation[3] = 0, rotation[4] = 1, rotation[5] = 0,
+        rotation[6] = 0, rotation[7] = 0, rotation[8] = 1;
+    if (scale) *scale = 1.0f;
+
+    __m512 const zeros_f32x16 = _mm512_setzero_ps();
+    __m512 sum_a_x_f32x16 = zeros_f32x16, sum_a_y_f32x16 = zeros_f32x16, sum_a_z_f32x16 = zeros_f32x16;
+    __m512 sum_b_x_f32x16 = zeros_f32x16, sum_b_y_f32x16 = zeros_f32x16, sum_b_z_f32x16 = zeros_f32x16;
+    __m512 sum_squared_x_f32x16 = zeros_f32x16, sum_squared_y_f32x16 = zeros_f32x16;
+    __m512 sum_squared_z_f32x16 = zeros_f32x16;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+    nk_size_t i = 0;
+
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(a_x_f32x16, b_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(a_y_f32x16, b_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(a_z_f32x16, b_z_f32x16);
+
+        sum_squared_x_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_x_f32x16);
+        sum_squared_y_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_y_f32x16);
+        sum_squared_z_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_z_f32x16);
+    }
+
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (i < n) {
+        nk_size_t tail = n - i;
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(a + i * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(b + i * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        __m512 delta_x_f32x16 = _mm512_sub_ps(a_x_f32x16, b_x_f32x16);
+        __m512 delta_y_f32x16 = _mm512_sub_ps(a_y_f32x16, b_y_f32x16);
+        __m512 delta_z_f32x16 = _mm512_sub_ps(a_z_f32x16, b_z_f32x16);
+
+        sum_squared_x_f32x16 = _mm512_fmadd_ps(delta_x_f32x16, delta_x_f32x16, sum_squared_x_f32x16);
+        sum_squared_y_f32x16 = _mm512_fmadd_ps(delta_y_f32x16, delta_y_f32x16, sum_squared_y_f32x16);
+        sum_squared_z_f32x16 = _mm512_fmadd_ps(delta_z_f32x16, delta_z_f32x16, sum_squared_z_f32x16);
+    }
+
+    nk_f32_t total_ax = _mm512_reduce_add_ps(sum_a_x_f32x16);
+    nk_f32_t total_ay = _mm512_reduce_add_ps(sum_a_y_f32x16);
+    nk_f32_t total_az = _mm512_reduce_add_ps(sum_a_z_f32x16);
+    nk_f32_t total_bx = _mm512_reduce_add_ps(sum_b_x_f32x16);
+    nk_f32_t total_by = _mm512_reduce_add_ps(sum_b_y_f32x16);
+    nk_f32_t total_bz = _mm512_reduce_add_ps(sum_b_z_f32x16);
+    nk_f32_t total_sq_x = _mm512_reduce_add_ps(sum_squared_x_f32x16);
+    nk_f32_t total_sq_y = _mm512_reduce_add_ps(sum_squared_y_f32x16);
+    nk_f32_t total_sq_z = _mm512_reduce_add_ps(sum_squared_z_f32x16);
+
+    nk_f32_t inv_n = 1.0f / (nk_f32_t)n;
+    nk_f32_t centroid_a_x = total_ax * inv_n;
+    nk_f32_t centroid_a_y = total_ay * inv_n;
+    nk_f32_t centroid_a_z = total_az * inv_n;
+    nk_f32_t centroid_b_x = total_bx * inv_n;
+    nk_f32_t centroid_b_y = total_by * inv_n;
+    nk_f32_t centroid_b_z = total_bz * inv_n;
+
+    if (a_centroid) a_centroid[0] = centroid_a_x, a_centroid[1] = centroid_a_y, a_centroid[2] = centroid_a_z;
+    if (b_centroid) b_centroid[0] = centroid_b_x, b_centroid[1] = centroid_b_y, b_centroid[2] = centroid_b_z;
+
+    nk_f32_t mean_diff_x = centroid_a_x - centroid_b_x;
+    nk_f32_t mean_diff_y = centroid_a_y - centroid_b_y;
+    nk_f32_t mean_diff_z = centroid_a_z - centroid_b_z;
+    nk_f32_t sum_squared = total_sq_x + total_sq_y + total_sq_z;
+    nk_f32_t mean_diff_sq = mean_diff_x * mean_diff_x + mean_diff_y * mean_diff_y + mean_diff_z * mean_diff_z;
+
+    *result = nk_f32_sqrt_haswell(sum_squared * inv_n - mean_diff_sq);
+}
+
+NK_PUBLIC void nk_kabsch_f16_skylake(nk_f16_t const *a, nk_f16_t const *b, nk_size_t n, nk_f32_t *a_centroid,
+                                     nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f32_t *result) {
+    __m512 const zeros_f32x16 = _mm512_setzero_ps();
+
+    __m512 sum_a_x_f32x16 = zeros_f32x16, sum_a_y_f32x16 = zeros_f32x16, sum_a_z_f32x16 = zeros_f32x16;
+    __m512 sum_b_x_f32x16 = zeros_f32x16, sum_b_y_f32x16 = zeros_f32x16, sum_b_z_f32x16 = zeros_f32x16;
+    __m512 cov_xx_f32x16 = zeros_f32x16, cov_xy_f32x16 = zeros_f32x16, cov_xz_f32x16 = zeros_f32x16;
+    __m512 cov_yx_f32x16 = zeros_f32x16, cov_yy_f32x16 = zeros_f32x16, cov_yz_f32x16 = zeros_f32x16;
+    __m512 cov_zx_f32x16 = zeros_f32x16, cov_zy_f32x16 = zeros_f32x16, cov_zz_f32x16 = zeros_f32x16;
+
+    nk_size_t i = 0;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_f16x16_to_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16x16_to_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+    }
+
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (i < n) {
+        nk_size_t tail = n - i;
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(a + i * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(b + i * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+    }
+
+    nk_f32_t sum_a_x = _mm512_reduce_add_ps(sum_a_x_f32x16);
+    nk_f32_t sum_a_y = _mm512_reduce_add_ps(sum_a_y_f32x16);
+    nk_f32_t sum_a_z = _mm512_reduce_add_ps(sum_a_z_f32x16);
+    nk_f32_t sum_b_x = _mm512_reduce_add_ps(sum_b_x_f32x16);
+    nk_f32_t sum_b_y = _mm512_reduce_add_ps(sum_b_y_f32x16);
+    nk_f32_t sum_b_z = _mm512_reduce_add_ps(sum_b_z_f32x16);
+    nk_f32_t covariance_x_x = _mm512_reduce_add_ps(cov_xx_f32x16);
+    nk_f32_t covariance_x_y = _mm512_reduce_add_ps(cov_xy_f32x16);
+    nk_f32_t covariance_x_z = _mm512_reduce_add_ps(cov_xz_f32x16);
+    nk_f32_t covariance_y_x = _mm512_reduce_add_ps(cov_yx_f32x16);
+    nk_f32_t covariance_y_y = _mm512_reduce_add_ps(cov_yy_f32x16);
+    nk_f32_t covariance_y_z = _mm512_reduce_add_ps(cov_yz_f32x16);
+    nk_f32_t covariance_z_x = _mm512_reduce_add_ps(cov_zx_f32x16);
+    nk_f32_t covariance_z_y = _mm512_reduce_add_ps(cov_zy_f32x16);
+    nk_f32_t covariance_z_z = _mm512_reduce_add_ps(cov_zz_f32x16);
+
+    nk_f32_t inv_n = 1.0f / (nk_f32_t)n;
+    nk_f32_t centroid_a_x = sum_a_x * inv_n, centroid_a_y = sum_a_y * inv_n, centroid_a_z = sum_a_z * inv_n;
+    nk_f32_t centroid_b_x = sum_b_x * inv_n, centroid_b_y = sum_b_y * inv_n, centroid_b_z = sum_b_z * inv_n;
+
+    if (a_centroid) a_centroid[0] = centroid_a_x, a_centroid[1] = centroid_a_y, a_centroid[2] = centroid_a_z;
+    if (b_centroid) b_centroid[0] = centroid_b_x, b_centroid[1] = centroid_b_y, b_centroid[2] = centroid_b_z;
+
+    covariance_x_x -= (nk_f32_t)n * centroid_a_x * centroid_b_x;
+    covariance_x_y -= (nk_f32_t)n * centroid_a_x * centroid_b_y;
+    covariance_x_z -= (nk_f32_t)n * centroid_a_x * centroid_b_z;
+    covariance_y_x -= (nk_f32_t)n * centroid_a_y * centroid_b_x;
+    covariance_y_y -= (nk_f32_t)n * centroid_a_y * centroid_b_y;
+    covariance_y_z -= (nk_f32_t)n * centroid_a_y * centroid_b_z;
+    covariance_z_x -= (nk_f32_t)n * centroid_a_z * centroid_b_x;
+    covariance_z_y -= (nk_f32_t)n * centroid_a_z * centroid_b_y;
+    covariance_z_z -= (nk_f32_t)n * centroid_a_z * centroid_b_z;
+
+    nk_f32_t cross_covariance[9] = {covariance_x_x, covariance_x_y, covariance_x_z, covariance_y_x, covariance_y_y,
+                                    covariance_y_z, covariance_z_x, covariance_z_y, covariance_z_z};
+    nk_f32_t svd_u[9], svd_s[9], svd_v[9];
+    nk_svd3x3_f32_(cross_covariance, svd_u, svd_s, svd_v);
+
+    nk_f32_t r[9];
+    r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+    r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+    r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+    r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+    r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+    r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+    r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+    r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+    r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+
+    if (nk_det3x3_f32_(r) < 0) {
+        svd_v[2] = -svd_v[2], svd_v[5] = -svd_v[5], svd_v[8] = -svd_v[8];
+        r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+        r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+        r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+        r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+        r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+        r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+        r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+        r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+        r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+    }
+
+    if (rotation)
+        for (int j = 0; j < 9; ++j) rotation[j] = r[j];
+    if (scale) *scale = 1.0f;
+
+    nk_f32_t sum_squared = nk_transformed_ssd_f16_skylake_(a, b, n, r, 1.0f, centroid_a_x, centroid_a_y, centroid_a_z,
+                                                           centroid_b_x, centroid_b_y, centroid_b_z);
+    *result = nk_f32_sqrt_haswell(sum_squared * inv_n);
+}
+
+NK_PUBLIC void nk_kabsch_bf16_skylake(nk_bf16_t const *a, nk_bf16_t const *b, nk_size_t n, nk_f32_t *a_centroid,
+                                      nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f32_t *result) {
+    __m512 const zeros_f32x16 = _mm512_setzero_ps();
+
+    __m512 sum_a_x_f32x16 = zeros_f32x16, sum_a_y_f32x16 = zeros_f32x16, sum_a_z_f32x16 = zeros_f32x16;
+    __m512 sum_b_x_f32x16 = zeros_f32x16, sum_b_y_f32x16 = zeros_f32x16, sum_b_z_f32x16 = zeros_f32x16;
+    __m512 cov_xx_f32x16 = zeros_f32x16, cov_xy_f32x16 = zeros_f32x16, cov_xz_f32x16 = zeros_f32x16;
+    __m512 cov_yx_f32x16 = zeros_f32x16, cov_yy_f32x16 = zeros_f32x16, cov_yz_f32x16 = zeros_f32x16;
+    __m512 cov_zx_f32x16 = zeros_f32x16, cov_zy_f32x16 = zeros_f32x16, cov_zz_f32x16 = zeros_f32x16;
+
+    nk_size_t i = 0;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+    }
+
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (i < n) {
+        nk_size_t tail = n - i;
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(a + i * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(b + i * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+    }
+
+    nk_f32_t sum_a_x = _mm512_reduce_add_ps(sum_a_x_f32x16);
+    nk_f32_t sum_a_y = _mm512_reduce_add_ps(sum_a_y_f32x16);
+    nk_f32_t sum_a_z = _mm512_reduce_add_ps(sum_a_z_f32x16);
+    nk_f32_t sum_b_x = _mm512_reduce_add_ps(sum_b_x_f32x16);
+    nk_f32_t sum_b_y = _mm512_reduce_add_ps(sum_b_y_f32x16);
+    nk_f32_t sum_b_z = _mm512_reduce_add_ps(sum_b_z_f32x16);
+    nk_f32_t covariance_x_x = _mm512_reduce_add_ps(cov_xx_f32x16);
+    nk_f32_t covariance_x_y = _mm512_reduce_add_ps(cov_xy_f32x16);
+    nk_f32_t covariance_x_z = _mm512_reduce_add_ps(cov_xz_f32x16);
+    nk_f32_t covariance_y_x = _mm512_reduce_add_ps(cov_yx_f32x16);
+    nk_f32_t covariance_y_y = _mm512_reduce_add_ps(cov_yy_f32x16);
+    nk_f32_t covariance_y_z = _mm512_reduce_add_ps(cov_yz_f32x16);
+    nk_f32_t covariance_z_x = _mm512_reduce_add_ps(cov_zx_f32x16);
+    nk_f32_t covariance_z_y = _mm512_reduce_add_ps(cov_zy_f32x16);
+    nk_f32_t covariance_z_z = _mm512_reduce_add_ps(cov_zz_f32x16);
+
+    nk_f32_t inv_n = 1.0f / (nk_f32_t)n;
+    nk_f32_t centroid_a_x = sum_a_x * inv_n, centroid_a_y = sum_a_y * inv_n, centroid_a_z = sum_a_z * inv_n;
+    nk_f32_t centroid_b_x = sum_b_x * inv_n, centroid_b_y = sum_b_y * inv_n, centroid_b_z = sum_b_z * inv_n;
+
+    if (a_centroid) a_centroid[0] = centroid_a_x, a_centroid[1] = centroid_a_y, a_centroid[2] = centroid_a_z;
+    if (b_centroid) b_centroid[0] = centroid_b_x, b_centroid[1] = centroid_b_y, b_centroid[2] = centroid_b_z;
+
+    covariance_x_x -= (nk_f32_t)n * centroid_a_x * centroid_b_x;
+    covariance_x_y -= (nk_f32_t)n * centroid_a_x * centroid_b_y;
+    covariance_x_z -= (nk_f32_t)n * centroid_a_x * centroid_b_z;
+    covariance_y_x -= (nk_f32_t)n * centroid_a_y * centroid_b_x;
+    covariance_y_y -= (nk_f32_t)n * centroid_a_y * centroid_b_y;
+    covariance_y_z -= (nk_f32_t)n * centroid_a_y * centroid_b_z;
+    covariance_z_x -= (nk_f32_t)n * centroid_a_z * centroid_b_x;
+    covariance_z_y -= (nk_f32_t)n * centroid_a_z * centroid_b_y;
+    covariance_z_z -= (nk_f32_t)n * centroid_a_z * centroid_b_z;
+
+    nk_f32_t cross_covariance[9] = {covariance_x_x, covariance_x_y, covariance_x_z, covariance_y_x, covariance_y_y,
+                                    covariance_y_z, covariance_z_x, covariance_z_y, covariance_z_z};
+    nk_f32_t svd_u[9], svd_s[9], svd_v[9];
+    nk_svd3x3_f32_(cross_covariance, svd_u, svd_s, svd_v);
+
+    nk_f32_t r[9];
+    r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+    r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+    r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+    r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+    r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+    r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+    r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+    r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+    r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+
+    if (nk_det3x3_f32_(r) < 0) {
+        svd_v[2] = -svd_v[2], svd_v[5] = -svd_v[5], svd_v[8] = -svd_v[8];
+        r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+        r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+        r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+        r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+        r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+        r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+        r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+        r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+        r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+    }
+
+    if (rotation)
+        for (int j = 0; j < 9; ++j) rotation[j] = r[j];
+    if (scale) *scale = 1.0f;
+
+    nk_f32_t sum_squared = nk_transformed_ssd_bf16_skylake_(a, b, n, r, 1.0f, centroid_a_x, centroid_a_y, centroid_a_z,
+                                                            centroid_b_x, centroid_b_y, centroid_b_z);
+    *result = nk_f32_sqrt_haswell(sum_squared * inv_n);
+}
+
+NK_PUBLIC void nk_umeyama_f16_skylake(nk_f16_t const *a, nk_f16_t const *b, nk_size_t n, nk_f32_t *a_centroid,
+                                      nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f32_t *result) {
+    __m512 const zeros_f32x16 = _mm512_setzero_ps();
+
+    __m512 sum_a_x_f32x16 = zeros_f32x16, sum_a_y_f32x16 = zeros_f32x16, sum_a_z_f32x16 = zeros_f32x16;
+    __m512 sum_b_x_f32x16 = zeros_f32x16, sum_b_y_f32x16 = zeros_f32x16, sum_b_z_f32x16 = zeros_f32x16;
+    __m512 cov_xx_f32x16 = zeros_f32x16, cov_xy_f32x16 = zeros_f32x16, cov_xz_f32x16 = zeros_f32x16;
+    __m512 cov_yx_f32x16 = zeros_f32x16, cov_yy_f32x16 = zeros_f32x16, cov_yz_f32x16 = zeros_f32x16;
+    __m512 cov_zx_f32x16 = zeros_f32x16, cov_zy_f32x16 = zeros_f32x16, cov_zz_f32x16 = zeros_f32x16;
+    __m512 variance_a_f32x16 = zeros_f32x16;
+
+    nk_size_t i = 0;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_f16x16_to_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16x16_to_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+
+        variance_a_f32x16 = _mm512_fmadd_ps(a_x_f32x16, a_x_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_y_f32x16, a_y_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_z_f32x16, a_z_f32x16, variance_a_f32x16);
+    }
+
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (i < n) {
+        nk_size_t tail = n - i;
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(a + i * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_f16_tail_to_f32x16_skylake_(b + i * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+
+        variance_a_f32x16 = _mm512_fmadd_ps(a_x_f32x16, a_x_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_y_f32x16, a_y_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_z_f32x16, a_z_f32x16, variance_a_f32x16);
+    }
+
+    nk_f32_t sum_a_x = _mm512_reduce_add_ps(sum_a_x_f32x16);
+    nk_f32_t sum_a_y = _mm512_reduce_add_ps(sum_a_y_f32x16);
+    nk_f32_t sum_a_z = _mm512_reduce_add_ps(sum_a_z_f32x16);
+    nk_f32_t sum_b_x = _mm512_reduce_add_ps(sum_b_x_f32x16);
+    nk_f32_t sum_b_y = _mm512_reduce_add_ps(sum_b_y_f32x16);
+    nk_f32_t sum_b_z = _mm512_reduce_add_ps(sum_b_z_f32x16);
+    nk_f32_t covariance_x_x = _mm512_reduce_add_ps(cov_xx_f32x16);
+    nk_f32_t covariance_x_y = _mm512_reduce_add_ps(cov_xy_f32x16);
+    nk_f32_t covariance_x_z = _mm512_reduce_add_ps(cov_xz_f32x16);
+    nk_f32_t covariance_y_x = _mm512_reduce_add_ps(cov_yx_f32x16);
+    nk_f32_t covariance_y_y = _mm512_reduce_add_ps(cov_yy_f32x16);
+    nk_f32_t covariance_y_z = _mm512_reduce_add_ps(cov_yz_f32x16);
+    nk_f32_t covariance_z_x = _mm512_reduce_add_ps(cov_zx_f32x16);
+    nk_f32_t covariance_z_y = _mm512_reduce_add_ps(cov_zy_f32x16);
+    nk_f32_t covariance_z_z = _mm512_reduce_add_ps(cov_zz_f32x16);
+    nk_f32_t variance_a_sum = _mm512_reduce_add_ps(variance_a_f32x16);
+
+    nk_f32_t inv_n = 1.0f / (nk_f32_t)n;
+    nk_f32_t centroid_a_x = sum_a_x * inv_n, centroid_a_y = sum_a_y * inv_n, centroid_a_z = sum_a_z * inv_n;
+    nk_f32_t centroid_b_x = sum_b_x * inv_n, centroid_b_y = sum_b_y * inv_n, centroid_b_z = sum_b_z * inv_n;
+
+    if (a_centroid) a_centroid[0] = centroid_a_x, a_centroid[1] = centroid_a_y, a_centroid[2] = centroid_a_z;
+    if (b_centroid) b_centroid[0] = centroid_b_x, b_centroid[1] = centroid_b_y, b_centroid[2] = centroid_b_z;
+
+    nk_f32_t variance_a = variance_a_sum * inv_n -
+                          (centroid_a_x * centroid_a_x + centroid_a_y * centroid_a_y + centroid_a_z * centroid_a_z);
+
+    covariance_x_x -= (nk_f32_t)n * centroid_a_x * centroid_b_x;
+    covariance_x_y -= (nk_f32_t)n * centroid_a_x * centroid_b_y;
+    covariance_x_z -= (nk_f32_t)n * centroid_a_x * centroid_b_z;
+    covariance_y_x -= (nk_f32_t)n * centroid_a_y * centroid_b_x;
+    covariance_y_y -= (nk_f32_t)n * centroid_a_y * centroid_b_y;
+    covariance_y_z -= (nk_f32_t)n * centroid_a_y * centroid_b_z;
+    covariance_z_x -= (nk_f32_t)n * centroid_a_z * centroid_b_x;
+    covariance_z_y -= (nk_f32_t)n * centroid_a_z * centroid_b_y;
+    covariance_z_z -= (nk_f32_t)n * centroid_a_z * centroid_b_z;
+
+    nk_f32_t cross_covariance[9] = {covariance_x_x, covariance_x_y, covariance_x_z, covariance_y_x, covariance_y_y,
+                                    covariance_y_z, covariance_z_x, covariance_z_y, covariance_z_z};
+
+    nk_f32_t svd_u[9], svd_s[9], svd_v[9];
+    nk_svd3x3_f32_(cross_covariance, svd_u, svd_s, svd_v);
+
+    nk_f32_t r[9];
+    r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+    r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+    r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+    r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+    r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+    r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+    r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+    r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+    r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+
+    nk_f32_t det = nk_det3x3_f32_(r);
+    nk_f32_t d3 = det < 0 ? -1.0f : 1.0f;
+    nk_f32_t trace_ds = svd_s[0] + svd_s[4] + d3 * svd_s[8];
+    nk_f32_t c = trace_ds / ((nk_f32_t)n * variance_a);
+    if (scale) *scale = c;
+
+    if (det < 0) {
+        svd_v[2] = -svd_v[2], svd_v[5] = -svd_v[5], svd_v[8] = -svd_v[8];
+        r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+        r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+        r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+        r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+        r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+        r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+        r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+        r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+        r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+    }
+
+    if (rotation)
+        for (int j = 0; j < 9; ++j) rotation[j] = r[j];
+
+    nk_f32_t sum_squared = nk_transformed_ssd_f16_skylake_(a, b, n, r, c, centroid_a_x, centroid_a_y, centroid_a_z,
+                                                           centroid_b_x, centroid_b_y, centroid_b_z);
+    *result = nk_f32_sqrt_haswell(sum_squared * inv_n);
+}
+
+NK_PUBLIC void nk_umeyama_bf16_skylake(nk_bf16_t const *a, nk_bf16_t const *b, nk_size_t n, nk_f32_t *a_centroid,
+                                       nk_f32_t *b_centroid, nk_f32_t *rotation, nk_f32_t *scale, nk_f32_t *result) {
+    __m512 const zeros_f32x16 = _mm512_setzero_ps();
+
+    __m512 sum_a_x_f32x16 = zeros_f32x16, sum_a_y_f32x16 = zeros_f32x16, sum_a_z_f32x16 = zeros_f32x16;
+    __m512 sum_b_x_f32x16 = zeros_f32x16, sum_b_y_f32x16 = zeros_f32x16, sum_b_z_f32x16 = zeros_f32x16;
+    __m512 cov_xx_f32x16 = zeros_f32x16, cov_xy_f32x16 = zeros_f32x16, cov_xz_f32x16 = zeros_f32x16;
+    __m512 cov_yx_f32x16 = zeros_f32x16, cov_yy_f32x16 = zeros_f32x16, cov_yz_f32x16 = zeros_f32x16;
+    __m512 cov_zx_f32x16 = zeros_f32x16, cov_zy_f32x16 = zeros_f32x16, cov_zz_f32x16 = zeros_f32x16;
+    __m512 variance_a_f32x16 = zeros_f32x16;
+
+    nk_size_t i = 0;
+    __m512 a_x_f32x16, a_y_f32x16, a_z_f32x16, b_x_f32x16, b_y_f32x16, b_z_f32x16;
+
+    for (; i + 16 <= n; i += 16) {
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(a + i * 3, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16x16_to_f32x16_skylake_(b + i * 3, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+
+        variance_a_f32x16 = _mm512_fmadd_ps(a_x_f32x16, a_x_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_y_f32x16, a_y_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_z_f32x16, a_z_f32x16, variance_a_f32x16);
+    }
+
+    // Tail: deinterleave remaining points into zero-initialized vectors
+    if (i < n) {
+        nk_size_t tail = n - i;
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(a + i * 3, tail, &a_x_f32x16, &a_y_f32x16, &a_z_f32x16);
+        nk_deinterleave_bf16_tail_to_f32x16_skylake_(b + i * 3, tail, &b_x_f32x16, &b_y_f32x16, &b_z_f32x16);
+
+        sum_a_x_f32x16 = _mm512_add_ps(sum_a_x_f32x16, a_x_f32x16);
+        sum_a_y_f32x16 = _mm512_add_ps(sum_a_y_f32x16, a_y_f32x16);
+        sum_a_z_f32x16 = _mm512_add_ps(sum_a_z_f32x16, a_z_f32x16);
+        sum_b_x_f32x16 = _mm512_add_ps(sum_b_x_f32x16, b_x_f32x16);
+        sum_b_y_f32x16 = _mm512_add_ps(sum_b_y_f32x16, b_y_f32x16);
+        sum_b_z_f32x16 = _mm512_add_ps(sum_b_z_f32x16, b_z_f32x16);
+
+        cov_xx_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_x_f32x16, cov_xx_f32x16);
+        cov_xy_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_y_f32x16, cov_xy_f32x16);
+        cov_xz_f32x16 = _mm512_fmadd_ps(a_x_f32x16, b_z_f32x16, cov_xz_f32x16);
+        cov_yx_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_x_f32x16, cov_yx_f32x16);
+        cov_yy_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_y_f32x16, cov_yy_f32x16);
+        cov_yz_f32x16 = _mm512_fmadd_ps(a_y_f32x16, b_z_f32x16, cov_yz_f32x16);
+        cov_zx_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_x_f32x16, cov_zx_f32x16);
+        cov_zy_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_y_f32x16, cov_zy_f32x16);
+        cov_zz_f32x16 = _mm512_fmadd_ps(a_z_f32x16, b_z_f32x16, cov_zz_f32x16);
+
+        variance_a_f32x16 = _mm512_fmadd_ps(a_x_f32x16, a_x_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_y_f32x16, a_y_f32x16, variance_a_f32x16);
+        variance_a_f32x16 = _mm512_fmadd_ps(a_z_f32x16, a_z_f32x16, variance_a_f32x16);
+    }
+
+    nk_f32_t sum_a_x = _mm512_reduce_add_ps(sum_a_x_f32x16);
+    nk_f32_t sum_a_y = _mm512_reduce_add_ps(sum_a_y_f32x16);
+    nk_f32_t sum_a_z = _mm512_reduce_add_ps(sum_a_z_f32x16);
+    nk_f32_t sum_b_x = _mm512_reduce_add_ps(sum_b_x_f32x16);
+    nk_f32_t sum_b_y = _mm512_reduce_add_ps(sum_b_y_f32x16);
+    nk_f32_t sum_b_z = _mm512_reduce_add_ps(sum_b_z_f32x16);
+    nk_f32_t covariance_x_x = _mm512_reduce_add_ps(cov_xx_f32x16);
+    nk_f32_t covariance_x_y = _mm512_reduce_add_ps(cov_xy_f32x16);
+    nk_f32_t covariance_x_z = _mm512_reduce_add_ps(cov_xz_f32x16);
+    nk_f32_t covariance_y_x = _mm512_reduce_add_ps(cov_yx_f32x16);
+    nk_f32_t covariance_y_y = _mm512_reduce_add_ps(cov_yy_f32x16);
+    nk_f32_t covariance_y_z = _mm512_reduce_add_ps(cov_yz_f32x16);
+    nk_f32_t covariance_z_x = _mm512_reduce_add_ps(cov_zx_f32x16);
+    nk_f32_t covariance_z_y = _mm512_reduce_add_ps(cov_zy_f32x16);
+    nk_f32_t covariance_z_z = _mm512_reduce_add_ps(cov_zz_f32x16);
+    nk_f32_t variance_a_sum = _mm512_reduce_add_ps(variance_a_f32x16);
+
+    nk_f32_t inv_n = 1.0f / (nk_f32_t)n;
+    nk_f32_t centroid_a_x = sum_a_x * inv_n, centroid_a_y = sum_a_y * inv_n, centroid_a_z = sum_a_z * inv_n;
+    nk_f32_t centroid_b_x = sum_b_x * inv_n, centroid_b_y = sum_b_y * inv_n, centroid_b_z = sum_b_z * inv_n;
+
+    if (a_centroid) a_centroid[0] = centroid_a_x, a_centroid[1] = centroid_a_y, a_centroid[2] = centroid_a_z;
+    if (b_centroid) b_centroid[0] = centroid_b_x, b_centroid[1] = centroid_b_y, b_centroid[2] = centroid_b_z;
+
+    nk_f32_t variance_a = variance_a_sum * inv_n -
+                          (centroid_a_x * centroid_a_x + centroid_a_y * centroid_a_y + centroid_a_z * centroid_a_z);
+
+    covariance_x_x -= (nk_f32_t)n * centroid_a_x * centroid_b_x;
+    covariance_x_y -= (nk_f32_t)n * centroid_a_x * centroid_b_y;
+    covariance_x_z -= (nk_f32_t)n * centroid_a_x * centroid_b_z;
+    covariance_y_x -= (nk_f32_t)n * centroid_a_y * centroid_b_x;
+    covariance_y_y -= (nk_f32_t)n * centroid_a_y * centroid_b_y;
+    covariance_y_z -= (nk_f32_t)n * centroid_a_y * centroid_b_z;
+    covariance_z_x -= (nk_f32_t)n * centroid_a_z * centroid_b_x;
+    covariance_z_y -= (nk_f32_t)n * centroid_a_z * centroid_b_y;
+    covariance_z_z -= (nk_f32_t)n * centroid_a_z * centroid_b_z;
+
+    nk_f32_t cross_covariance[9] = {covariance_x_x, covariance_x_y, covariance_x_z, covariance_y_x, covariance_y_y,
+                                    covariance_y_z, covariance_z_x, covariance_z_y, covariance_z_z};
+
+    nk_f32_t svd_u[9], svd_s[9], svd_v[9];
+    nk_svd3x3_f32_(cross_covariance, svd_u, svd_s, svd_v);
+
+    nk_f32_t r[9];
+    r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+    r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+    r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+    r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+    r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+    r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+    r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+    r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+    r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+
+    nk_f32_t det = nk_det3x3_f32_(r);
+    nk_f32_t d3 = det < 0 ? -1.0f : 1.0f;
+    nk_f32_t trace_ds = svd_s[0] + svd_s[4] + d3 * svd_s[8];
+    nk_f32_t c = trace_ds / ((nk_f32_t)n * variance_a);
+    if (scale) *scale = c;
+
+    if (det < 0) {
+        svd_v[2] = -svd_v[2], svd_v[5] = -svd_v[5], svd_v[8] = -svd_v[8];
+        r[0] = svd_v[0] * svd_u[0] + svd_v[1] * svd_u[1] + svd_v[2] * svd_u[2];
+        r[1] = svd_v[0] * svd_u[3] + svd_v[1] * svd_u[4] + svd_v[2] * svd_u[5];
+        r[2] = svd_v[0] * svd_u[6] + svd_v[1] * svd_u[7] + svd_v[2] * svd_u[8];
+        r[3] = svd_v[3] * svd_u[0] + svd_v[4] * svd_u[1] + svd_v[5] * svd_u[2];
+        r[4] = svd_v[3] * svd_u[3] + svd_v[4] * svd_u[4] + svd_v[5] * svd_u[5];
+        r[5] = svd_v[3] * svd_u[6] + svd_v[4] * svd_u[7] + svd_v[5] * svd_u[8];
+        r[6] = svd_v[6] * svd_u[0] + svd_v[7] * svd_u[1] + svd_v[8] * svd_u[2];
+        r[7] = svd_v[6] * svd_u[3] + svd_v[7] * svd_u[4] + svd_v[8] * svd_u[5];
+        r[8] = svd_v[6] * svd_u[6] + svd_v[7] * svd_u[7] + svd_v[8] * svd_u[8];
+    }
+
+    if (rotation)
+        for (int j = 0; j < 9; ++j) rotation[j] = r[j];
+
+    nk_f32_t sum_squared = nk_transformed_ssd_bf16_skylake_(a, b, n, r, c, centroid_a_x, centroid_a_y, centroid_a_z,
+                                                            centroid_b_x, centroid_b_y, centroid_b_z);
+    *result = nk_f32_sqrt_haswell(sum_squared * inv_n);
+}
+
 #if defined(__clang__)
 #pragma clang attribute pop
 #elif defined(__GNUC__)