diff --git a/crates/core_simd/src/permute.rs b/crates/core_simd/src/permute.rs
index b27b0a9e141..dd63c69c63d 100644
--- a/crates/core_simd/src/permute.rs
+++ b/crates/core_simd/src/permute.rs
@@ -13,6 +13,115 @@ impl $name<$n> {
             pub fn shuffle<const IDX: [u32; $n]>(self, second: Self) -> Self {
                 unsafe { crate::intrinsics::$fn(self, second, IDX) }
             }
+
+            /// Reverse the order of the lanes in the vector.
+            #[inline]
+            pub fn reverse(self) -> Self {
+                const fn idx() -> [u32; $n] {
+                    let mut idx = [0u32; $n];
+                    let mut i = 0;
+                    while i < $n {
+                        idx[i] = ($n - i - 1) as u32;
+                        i += 1;
+                    }
+                    idx
+                }
+                self.shuffle::<{ idx() }>(self)
+            }
+
+            /// Interleave two vectors.
+            ///
+            /// Produces two vectors with lanes taken alternately from `self` and `other`.
+            ///
+            /// The first result contains the first `LANES / 2` lanes from `self` and `other`,
+            /// alternating, starting with the first lane of `self`.
+            ///
+            /// The second result contains the last `LANES / 2` lanes from `self` and `other`,
+            /// alternating, starting with the lane `LANES / 2` from the start of `self`.
+            ///
+            /// This particular permutation is efficient on many architectures.
+            ///
+            /// ```
+            /// # use core_simd::SimdU32;
+            /// let a = SimdU32::from_array([0, 1, 2, 3]);
+            /// let b = SimdU32::from_array([4, 5, 6, 7]);
+            /// let (x, y) = a.interleave(b);
+            /// assert_eq!(x.to_array(), [0, 4, 1, 5]);
+            /// assert_eq!(y.to_array(), [2, 6, 3, 7]);
+            /// ```
+            #[inline]
+            pub fn interleave(self, other: Self) -> (Self, Self) {
+                const fn lo() -> [u32; $n] {
+                    let mut idx = [0u32; $n];
+                    let mut i = 0;
+                    while i < $n {
+                        let offset = i / 2;
+                        idx[i] = if i % 2 == 0 {
+                            offset
+                        } else {
+                            $n + offset
+                        } as u32;
+                        i += 1;
+                    }
+                    idx
+                }
+                const fn hi() -> [u32; $n] {
+                    let mut idx = [0u32; $n];
+                    let mut i = 0;
+                    while i < $n {
+                        let offset = ($n + i) / 2;
+                        idx[i] = if i % 2 == 0 {
+                            offset
+                        } else {
+                            $n + offset
+                        } as u32;
+                        i += 1;
+                    }
+                    idx
+                }
+                (self.shuffle::<{ lo() }>(other), self.shuffle::<{ hi() }>(other))
+            }
+
+            /// Deinterleave two vectors.
+            ///
+            /// The first result takes every other lane of `self` and then `other`, starting with
+            /// the first lane.
+            ///
+            /// The second result takes every other lane of `self` and then `other`, starting with
+            /// the second lane.
+            ///
+            /// This particular permutation is efficient on many architectures.
+            ///
+            /// ```
+            /// # use core_simd::SimdU32;
+            /// let a = SimdU32::from_array([0, 4, 1, 5]);
+            /// let b = SimdU32::from_array([2, 6, 3, 7]);
+            /// let (x, y) = a.deinterleave(b);
+            /// assert_eq!(x.to_array(), [0, 1, 2, 3]);
+            /// assert_eq!(y.to_array(), [4, 5, 6, 7]);
+            /// ```
+            #[inline]
+            pub fn deinterleave(self, other: Self) -> (Self, Self) {
+                const fn even() -> [u32; $n] {
+                    let mut idx = [0u32; $n];
+                    let mut i = 0;
+                    while i < $n {
+                        idx[i] = 2 * i as u32;
+                        i += 1;
+                    }
+                    idx
+                }
+                const fn odd() -> [u32; $n] {
+                    let mut idx = [0u32; $n];
+                    let mut i = 0;
+                    while i < $n {
+                        idx[i] = 1 + 2 * i as u32;
+                        i += 1;
+                    }
+                    idx
+                }
+                (self.shuffle::<{ even() }>(other), self.shuffle::<{ odd() }>(other))
+            }
         }
     }
 }
diff --git a/crates/core_simd/tests/permute.rs b/crates/core_simd/tests/permute.rs
index 1c6c391d8d0..2be43c9cf3c 100644
--- a/crates/core_simd/tests/permute.rs
+++ b/crates/core_simd/tests/permute.rs
@@ -13,3 +13,23 @@ fn simple_shuffle() {
     let b = a;
     assert_eq!(a.shuffle::<{ [3, 1, 4, 6] }>(b).to_array(), [9, 4, 2, 1]);
 }
+
+#[test]
+#[cfg_attr(target_arch = "wasm32", wasm_bindgen_test)]
+fn reverse() {
+    let a = SimdU32::from_array([0, 1, 2, 3, 4, 5, 6, 7]);
+    assert_eq!(a.reverse().to_array(), [7, 6, 5, 4, 3, 2, 1, 0]);
+}
+
+#[test]
+#[cfg_attr(target_arch = "wasm32", wasm_bindgen_test)]
+fn interleave() {
+    let a = SimdU32::from_array([0, 1, 2, 3, 4, 5, 6, 7]);
+    let b = SimdU32::from_array([8, 9, 10, 11, 12, 13, 14, 15]);
+    let (lo, hi) = a.interleave(b);
+    assert_eq!(lo.to_array(), [0, 8, 1, 9, 2, 10, 3, 11]);
+    assert_eq!(hi.to_array(), [4, 12, 5, 13, 6, 14, 7, 15]);
+    let (even, odd) = lo.deinterleave(hi);
+    assert_eq!(even, a);
+    assert_eq!(odd, b);
+}