diff --git a/crates/bevy_app/src/app.rs b/crates/bevy_app/src/app.rs
index 07a29800f93df..da41efbc152c4 100644
--- a/crates/bevy_app/src/app.rs
+++ b/crates/bevy_app/src/app.rs
@@ -18,7 +18,7 @@ use std::{
     process::{ExitCode, Termination},
 };
 use std::{
-    num::NonZeroU8,
+    num::NonZero,
     panic::{catch_unwind, resume_unwind, AssertUnwindSafe},
 };
 use thiserror::Error;
@@ -1061,14 +1061,14 @@ pub enum AppExit {
     Success,
     /// The [`App`] experienced an unhandleable error.
     /// Holds the exit code we expect our app to return.
-    Error(NonZeroU8),
+    Error(NonZero<u8>),
 }
 
 impl AppExit {
     /// Creates a [`AppExit::Error`] with a error code of 1.
     #[must_use]
     pub const fn error() -> Self {
-        Self::Error(NonZeroU8::MIN)
+        Self::Error(NonZero::<u8>::MIN)
     }
 
     /// Returns `true` if `self` is a [`AppExit::Success`].
@@ -1089,7 +1089,7 @@ impl AppExit {
     /// [`AppExit::Error`] is constructed.
     #[must_use]
     pub const fn from_code(code: u8) -> Self {
-        match NonZeroU8::new(code) {
+        match NonZero::<u8>::new(code) {
             Some(code) => Self::Error(code),
             None => Self::Success,
         }
diff --git a/crates/bevy_core_pipeline/src/auto_exposure/pipeline.rs b/crates/bevy_core_pipeline/src/auto_exposure/pipeline.rs
index eacff931c7211..937e18f410485 100644
--- a/crates/bevy_core_pipeline/src/auto_exposure/pipeline.rs
+++ b/crates/bevy_core_pipeline/src/auto_exposure/pipeline.rs
@@ -10,7 +10,7 @@ use bevy_render::{
     texture::Image,
     view::ViewUniform,
 };
-use std::num::NonZeroU64;
+use std::num::NonZero;
 
 #[derive(Resource)]
 pub struct AutoExposurePipeline {
@@ -64,8 +64,8 @@ impl FromWorld for AutoExposurePipeline {
                         texture_2d(TextureSampleType::Float { filterable: false }),
                         texture_1d(TextureSampleType::Float { filterable: false }),
                         uniform_buffer::<AutoExposureCompensationCurveUniform>(false),
-                        storage_buffer_sized(false, NonZeroU64::new(HISTOGRAM_BIN_COUNT * 4)),
-                        storage_buffer_sized(false, NonZeroU64::new(4)),
+                        storage_buffer_sized(false, NonZero::<u64>::new(HISTOGRAM_BIN_COUNT * 4)),
+                        storage_buffer_sized(false, NonZero::<u64>::new(4)),
                         storage_buffer::<ViewUniform>(true),
                     ),
                 ),
diff --git a/crates/bevy_ecs/src/entity/mod.rs b/crates/bevy_ecs/src/entity/mod.rs
index deee320fcd0b2..93b1e78ffabe3 100644
--- a/crates/bevy_ecs/src/entity/mod.rs
+++ b/crates/bevy_ecs/src/entity/mod.rs
@@ -59,7 +59,7 @@ use crate::{
 };
 #[cfg(feature = "serialize")]
 use serde::{Deserialize, Serialize};
-use std::{fmt, hash::Hash, mem, num::NonZeroU32, sync::atomic::Ordering};
+use std::{fmt, hash::Hash, mem, num::NonZero, sync::atomic::Ordering};
 
 #[cfg(target_has_atomic = "64")]
 use std::sync::atomic::AtomicI64 as AtomicIdCursor;
@@ -157,7 +157,7 @@ pub struct Entity {
     // to make this struct equivalent to a u64.
     #[cfg(target_endian = "little")]
     index: u32,
-    generation: NonZeroU32,
+    generation: NonZero<u32>,
     #[cfg(target_endian = "big")]
     index: u32,
 }
@@ -223,7 +223,7 @@ impl Entity {
     /// Construct an [`Entity`] from a raw `index` value and a non-zero `generation` value.
     /// Ensure that the generation value is never greater than `0x7FFF_FFFF`.
     #[inline(always)]
-    pub(crate) const fn from_raw_and_generation(index: u32, generation: NonZeroU32) -> Entity {
+    pub(crate) const fn from_raw_and_generation(index: u32, generation: NonZero<u32>) -> Entity {
         debug_assert!(generation.get() <= HIGH_MASK);
 
         Self { index, generation }
@@ -279,7 +279,7 @@ impl Entity {
     /// a component.
     #[inline(always)]
     pub const fn from_raw(index: u32) -> Entity {
-        Self::from_raw_and_generation(index, NonZeroU32::MIN)
+        Self::from_raw_and_generation(index, NonZero::<u32>::MIN)
     }
 
     /// Convert to a form convenient for passing outside of rust.
@@ -722,7 +722,7 @@ impl Entities {
 
         meta.generation = IdentifierMask::inc_masked_high_by(meta.generation, 1);
 
-        if meta.generation == NonZeroU32::MIN {
+        if meta.generation == NonZero::<u32>::MIN {
             warn!(
                 "Entity({}) generation wrapped on Entities::free, aliasing may occur",
                 entity.index
@@ -949,7 +949,7 @@ impl Entities {
 #[repr(C)]
 struct EntityMeta {
     /// The current generation of the [`Entity`].
-    pub generation: NonZeroU32,
+    pub generation: NonZero<u32>,
     /// The current location of the [`Entity`]
     pub location: EntityLocation,
 }
@@ -957,7 +957,7 @@ struct EntityMeta {
 impl EntityMeta {
     /// meta for **pending entity**
     const EMPTY: EntityMeta = EntityMeta {
-        generation: NonZeroU32::MIN,
+        generation: NonZero::<u32>::MIN,
         location: EntityLocation::INVALID,
     };
 }
@@ -1014,7 +1014,8 @@ mod tests {
     #[test]
     fn entity_bits_roundtrip() {
         // Generation cannot be greater than 0x7FFF_FFFF else it will be an invalid Entity id
-        let e = Entity::from_raw_and_generation(0xDEADBEEF, NonZeroU32::new(0x5AADF00D).unwrap());
+        let e =
+            Entity::from_raw_and_generation(0xDEADBEEF, NonZero::<u32>::new(0x5AADF00D).unwrap());
         assert_eq!(Entity::from_bits(e.to_bits()), e);
     }
 
@@ -1091,65 +1092,65 @@ mod tests {
     #[allow(clippy::nonminimal_bool)] // This is intentionally testing `lt` and `ge` as separate functions.
     fn entity_comparison() {
         assert_eq!(
-            Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap()),
-            Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap()),
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
         );
         assert_ne!(
-            Entity::from_raw_and_generation(123, NonZeroU32::new(789).unwrap()),
-            Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(789).unwrap()),
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
         );
         assert_ne!(
-            Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap()),
-            Entity::from_raw_and_generation(123, NonZeroU32::new(789).unwrap())
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap()),
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(789).unwrap())
         );
         assert_ne!(
-            Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap()),
-            Entity::from_raw_and_generation(456, NonZeroU32::new(123).unwrap())
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap()),
+            Entity::from_raw_and_generation(456, NonZero::<u32>::new(123).unwrap())
         );
 
         // ordering is by generation then by index
 
         assert!(
-            Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
-                >= Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
+                >= Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
         );
         assert!(
-            Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
-                <= Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
+            Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
+                <= Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
         );
         assert!(
-            !(Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
-                < Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap()))
+            !(Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
+                < Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap()))
         );
         assert!(
-            !(Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap())
-                > Entity::from_raw_and_generation(123, NonZeroU32::new(456).unwrap()))
+            !(Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap())
+                > Entity::from_raw_and_generation(123, NonZero::<u32>::new(456).unwrap()))
         );
 
         assert!(
-            Entity::from_raw_and_generation(9, NonZeroU32::new(1).unwrap())
-                < Entity::from_raw_and_generation(1, NonZeroU32::new(9).unwrap())
+            Entity::from_raw_and_generation(9, NonZero::<u32>::new(1).unwrap())
+                < Entity::from_raw_and_generation(1, NonZero::<u32>::new(9).unwrap())
         );
         assert!(
-            Entity::from_raw_and_generation(1, NonZeroU32::new(9).unwrap())
-                > Entity::from_raw_and_generation(9, NonZeroU32::new(1).unwrap())
+            Entity::from_raw_and_generation(1, NonZero::<u32>::new(9).unwrap())
+                > Entity::from_raw_and_generation(9, NonZero::<u32>::new(1).unwrap())
         );
 
         assert!(
-            Entity::from_raw_and_generation(1, NonZeroU32::new(1).unwrap())
-                < Entity::from_raw_and_generation(2, NonZeroU32::new(1).unwrap())
+            Entity::from_raw_and_generation(1, NonZero::<u32>::new(1).unwrap())
+                < Entity::from_raw_and_generation(2, NonZero::<u32>::new(1).unwrap())
         );
         assert!(
-            Entity::from_raw_and_generation(1, NonZeroU32::new(1).unwrap())
-                <= Entity::from_raw_and_generation(2, NonZeroU32::new(1).unwrap())
+            Entity::from_raw_and_generation(1, NonZero::<u32>::new(1).unwrap())
+                <= Entity::from_raw_and_generation(2, NonZero::<u32>::new(1).unwrap())
         );
         assert!(
-            Entity::from_raw_and_generation(2, NonZeroU32::new(2).unwrap())
-                > Entity::from_raw_and_generation(1, NonZeroU32::new(2).unwrap())
+            Entity::from_raw_and_generation(2, NonZero::<u32>::new(2).unwrap())
+                > Entity::from_raw_and_generation(1, NonZero::<u32>::new(2).unwrap())
         );
         assert!(
-            Entity::from_raw_and_generation(2, NonZeroU32::new(2).unwrap())
-                >= Entity::from_raw_and_generation(1, NonZeroU32::new(2).unwrap())
+            Entity::from_raw_and_generation(2, NonZero::<u32>::new(2).unwrap())
+                >= Entity::from_raw_and_generation(1, NonZero::<u32>::new(2).unwrap())
         );
     }
 
diff --git a/crates/bevy_ecs/src/identifier/masks.rs b/crates/bevy_ecs/src/identifier/masks.rs
index d5adc856ddb88..85fb393cb6f13 100644
--- a/crates/bevy_ecs/src/identifier/masks.rs
+++ b/crates/bevy_ecs/src/identifier/masks.rs
@@ -1,4 +1,4 @@
-use std::num::NonZeroU32;
+use std::num::NonZero;
 
 use super::kinds::IdKind;
 
@@ -61,7 +61,7 @@ impl IdentifierMask {
     /// Will never be greater than [`HIGH_MASK`] or less than `1`, and increments are masked to
     /// never be greater than [`HIGH_MASK`].
     #[inline(always)]
-    pub(crate) const fn inc_masked_high_by(lhs: NonZeroU32, rhs: u32) -> NonZeroU32 {
+    pub(crate) const fn inc_masked_high_by(lhs: NonZero<u32>, rhs: u32) -> NonZero<u32> {
         let lo = (lhs.get() & HIGH_MASK).wrapping_add(rhs & HIGH_MASK);
         // Checks high 32 bit for whether we have overflowed 31 bits.
         let overflowed = lo >> 31;
@@ -70,7 +70,7 @@ impl IdentifierMask {
         // - Adding the overflow flag will offset overflows to start at 1 instead of 0
         // - The sum of `0x7FFF_FFFF` + `u32::MAX` + 1 (overflow) == `0x7FFF_FFFF`
         // - If the operation doesn't overflow at 31 bits, no offsetting takes place
-        unsafe { NonZeroU32::new_unchecked(lo.wrapping_add(overflowed) & HIGH_MASK) }
+        unsafe { NonZero::<u32>::new_unchecked(lo.wrapping_add(overflowed) & HIGH_MASK) }
     }
 }
 
@@ -166,68 +166,68 @@ mod tests {
         // Adding from lowest value with lowest to highest increment
         // No result should ever be greater than 0x7FFF_FFFF or HIGH_MASK
         assert_eq!(
-            NonZeroU32::MIN,
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MIN, 0)
+            NonZero::<u32>::MIN,
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MIN, 0)
         );
         assert_eq!(
-            NonZeroU32::new(2).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MIN, 1)
+            NonZero::<u32>::new(2).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MIN, 1)
         );
         assert_eq!(
-            NonZeroU32::new(3).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MIN, 2)
+            NonZero::<u32>::new(3).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MIN, 2)
         );
         assert_eq!(
-            NonZeroU32::MIN,
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MIN, HIGH_MASK)
+            NonZero::<u32>::MIN,
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MIN, HIGH_MASK)
         );
         assert_eq!(
-            NonZeroU32::MIN,
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MIN, u32::MAX)
+            NonZero::<u32>::MIN,
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MIN, u32::MAX)
         );
         // Adding from absolute highest value with lowest to highest increment
         // No result should ever be greater than 0x7FFF_FFFF or HIGH_MASK
         assert_eq!(
-            NonZeroU32::new(HIGH_MASK).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MAX, 0)
+            NonZero::<u32>::new(HIGH_MASK).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MAX, 0)
         );
         assert_eq!(
-            NonZeroU32::MIN,
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MAX, 1)
+            NonZero::<u32>::MIN,
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MAX, 1)
         );
         assert_eq!(
-            NonZeroU32::new(2).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MAX, 2)
+            NonZero::<u32>::new(2).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MAX, 2)
         );
         assert_eq!(
-            NonZeroU32::new(HIGH_MASK).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MAX, HIGH_MASK)
+            NonZero::<u32>::new(HIGH_MASK).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MAX, HIGH_MASK)
         );
         assert_eq!(
-            NonZeroU32::new(HIGH_MASK).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::MAX, u32::MAX)
+            NonZero::<u32>::new(HIGH_MASK).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::MAX, u32::MAX)
         );
         // Adding from actual highest value with lowest to highest increment
         // No result should ever be greater than 0x7FFF_FFFF or HIGH_MASK
         assert_eq!(
-            NonZeroU32::new(HIGH_MASK).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::new(HIGH_MASK).unwrap(), 0)
+            NonZero::<u32>::new(HIGH_MASK).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::new(HIGH_MASK).unwrap(), 0)
         );
         assert_eq!(
-            NonZeroU32::MIN,
-            IdentifierMask::inc_masked_high_by(NonZeroU32::new(HIGH_MASK).unwrap(), 1)
+            NonZero::<u32>::MIN,
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::new(HIGH_MASK).unwrap(), 1)
         );
         assert_eq!(
-            NonZeroU32::new(2).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::new(HIGH_MASK).unwrap(), 2)
+            NonZero::<u32>::new(2).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::new(HIGH_MASK).unwrap(), 2)
         );
         assert_eq!(
-            NonZeroU32::new(HIGH_MASK).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::new(HIGH_MASK).unwrap(), HIGH_MASK)
+            NonZero::<u32>::new(HIGH_MASK).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::new(HIGH_MASK).unwrap(), HIGH_MASK)
         );
         assert_eq!(
-            NonZeroU32::new(HIGH_MASK).unwrap(),
-            IdentifierMask::inc_masked_high_by(NonZeroU32::new(HIGH_MASK).unwrap(), u32::MAX)
+            NonZero::<u32>::new(HIGH_MASK).unwrap(),
+            IdentifierMask::inc_masked_high_by(NonZero::<u32>::new(HIGH_MASK).unwrap(), u32::MAX)
         );
     }
 }
diff --git a/crates/bevy_ecs/src/identifier/mod.rs b/crates/bevy_ecs/src/identifier/mod.rs
index 04a93cde45c0c..e9c7df8006e38 100644
--- a/crates/bevy_ecs/src/identifier/mod.rs
+++ b/crates/bevy_ecs/src/identifier/mod.rs
@@ -7,7 +7,7 @@
 use bevy_reflect::Reflect;
 
 use self::{error::IdentifierError, kinds::IdKind, masks::IdentifierMask};
-use std::{hash::Hash, num::NonZeroU32};
+use std::{hash::Hash, num::NonZero};
 
 pub mod error;
 pub(crate) mod kinds;
@@ -28,7 +28,7 @@ pub struct Identifier {
     // to make this struct equivalent to a u64.
     #[cfg(target_endian = "little")]
     low: u32,
-    high: NonZeroU32,
+    high: NonZero<u32>,
     #[cfg(target_endian = "big")]
     low: u32,
 }
@@ -56,7 +56,7 @@ impl Identifier {
             unsafe {
                 Ok(Self {
                     low,
-                    high: NonZeroU32::new_unchecked(packed_high),
+                    high: NonZero::<u32>::new_unchecked(packed_high),
                 })
             }
         }
@@ -71,7 +71,7 @@ impl Identifier {
     /// Returns the value of the high segment of the [`Identifier`]. This
     /// does not apply any masking.
     #[inline(always)]
-    pub const fn high(self) -> NonZeroU32 {
+    pub const fn high(self) -> NonZero<u32> {
         self.high
     }
 
@@ -114,7 +114,7 @@ impl Identifier {
     /// This method is the fallible counterpart to [`Identifier::from_bits`].
     #[inline(always)]
     pub const fn try_from_bits(value: u64) -> Result<Self, IdentifierError> {
-        let high = NonZeroU32::new(IdentifierMask::get_high(value));
+        let high = NonZero::<u32>::new(IdentifierMask::get_high(value));
 
         match high {
             Some(high) => Ok(Self {
diff --git a/crates/bevy_ecs/src/lib.rs b/crates/bevy_ecs/src/lib.rs
index 09e86f7711cca..442c25d9a2662 100644
--- a/crates/bevy_ecs/src/lib.rs
+++ b/crates/bevy_ecs/src/lib.rs
@@ -88,7 +88,7 @@ mod tests {
     };
     use bevy_tasks::{ComputeTaskPool, TaskPool};
     use bevy_utils::HashSet;
-    use std::num::NonZeroU32;
+    use std::num::NonZero;
     use std::{
         any::TypeId,
         marker::PhantomData,
@@ -1659,7 +1659,7 @@ mod tests {
         );
 
         let e4_mismatched_generation =
-            Entity::from_raw_and_generation(3, NonZeroU32::new(2).unwrap());
+            Entity::from_raw_and_generation(3, NonZero::<u32>::new(2).unwrap());
         assert!(
             world_b.get_or_spawn(e4_mismatched_generation).is_none(),
             "attempting to spawn on top of an entity with a mismatched entity generation fails"
@@ -1754,7 +1754,8 @@ mod tests {
         let e0 = world.spawn(A(0)).id();
         let e1 = Entity::from_raw(1);
         let e2 = world.spawn_empty().id();
-        let invalid_e2 = Entity::from_raw_and_generation(e2.index(), NonZeroU32::new(2).unwrap());
+        let invalid_e2 =
+            Entity::from_raw_and_generation(e2.index(), NonZero::<u32>::new(2).unwrap());
 
         let values = vec![(e0, (B(0), C)), (e1, (B(1), C)), (invalid_e2, (B(2), C))];
 
diff --git a/crates/bevy_ecs/src/storage/blob_vec.rs b/crates/bevy_ecs/src/storage/blob_vec.rs
index dca9c1542a551..d5699f37164cb 100644
--- a/crates/bevy_ecs/src/storage/blob_vec.rs
+++ b/crates/bevy_ecs/src/storage/blob_vec.rs
@@ -1,7 +1,7 @@
 use std::{
     alloc::{handle_alloc_error, Layout},
     cell::UnsafeCell,
-    num::NonZeroUsize,
+    num::NonZero,
     ptr::NonNull,
 };
 
@@ -56,7 +56,7 @@ impl BlobVec {
         drop: Option<unsafe fn(OwningPtr<'_>)>,
         capacity: usize,
     ) -> BlobVec {
-        let align = NonZeroUsize::new(item_layout.align()).expect("alignment must be > 0");
+        let align = NonZero::<usize>::new(item_layout.align()).expect("alignment must be > 0");
         let data = bevy_ptr::dangling_with_align(align);
         if item_layout.size() == 0 {
             BlobVec {
@@ -119,7 +119,8 @@ impl BlobVec {
         let available_space = self.capacity - self.len;
         if available_space < additional {
             // SAFETY: `available_space < additional`, so `additional - available_space > 0`
-            let increment = unsafe { NonZeroUsize::new_unchecked(additional - available_space) };
+            let increment =
+                unsafe { NonZero::<usize>::new_unchecked(additional - available_space) };
             self.grow_exact(increment);
         }
     }
@@ -132,7 +133,7 @@ impl BlobVec {
         #[cold]
         fn do_reserve(slf: &mut BlobVec, additional: usize) {
             let increment = slf.capacity.max(additional - (slf.capacity - slf.len));
-            let increment = NonZeroUsize::new(increment).unwrap();
+            let increment = NonZero::<usize>::new(increment).unwrap();
             slf.grow_exact(increment);
         }
 
@@ -148,7 +149,7 @@ impl BlobVec {
     /// Panics if the new capacity overflows `usize`.
     /// For ZST it panics unconditionally because ZST `BlobVec` capacity
     /// is initialized to `usize::MAX` and always stays that way.
-    fn grow_exact(&mut self, increment: NonZeroUsize) {
+    fn grow_exact(&mut self, increment: NonZero<usize>) {
         let new_capacity = self
             .capacity
             .checked_add(increment.get())
diff --git a/crates/bevy_pbr/src/cluster/mod.rs b/crates/bevy_pbr/src/cluster/mod.rs
index fe913a1d196bb..7da6c5da026cf 100644
--- a/crates/bevy_pbr/src/cluster/mod.rs
+++ b/crates/bevy_pbr/src/cluster/mod.rs
@@ -1,6 +1,6 @@
 //! Spatial clustering of objects, currently just point and spot lights.
 
-use std::num::NonZeroU64;
+use std::num::NonZero;
 
 use bevy_core_pipeline::core_3d::Camera3d;
 use bevy_ecs::{
@@ -468,7 +468,7 @@ impl GpuClusterableObjects {
         }
     }
 
-    pub fn min_size(buffer_binding_type: BufferBindingType) -> NonZeroU64 {
+    pub fn min_size(buffer_binding_type: BufferBindingType) -> NonZero<u64> {
         match buffer_binding_type {
             BufferBindingType::Storage { .. } => GpuClusterableObjectsStorage::min_size(),
             BufferBindingType::Uniform => GpuClusterableObjectsUniform::min_size(),
@@ -749,7 +749,7 @@ impl ViewClusterBindings {
 
     pub fn min_size_clusterable_object_index_lists(
         buffer_binding_type: BufferBindingType,
-    ) -> NonZeroU64 {
+    ) -> NonZero<u64> {
         match buffer_binding_type {
             BufferBindingType::Storage { .. } => GpuClusterableObjectIndexListsStorage::min_size(),
             BufferBindingType::Uniform => GpuClusterableObjectIndexListsUniform::min_size(),
@@ -758,7 +758,7 @@ impl ViewClusterBindings {
 
     pub fn min_size_cluster_offsets_and_counts(
         buffer_binding_type: BufferBindingType,
-    ) -> NonZeroU64 {
+    ) -> NonZero<u64> {
         match buffer_binding_type {
             BufferBindingType::Storage { .. } => GpuClusterOffsetsAndCountsStorage::min_size(),
             BufferBindingType::Uniform => GpuClusterOffsetsAndCountsUniform::min_size(),
diff --git a/crates/bevy_pbr/src/light_probe/environment_map.rs b/crates/bevy_pbr/src/light_probe/environment_map.rs
index 8a78e93083024..1b1604df4d48d 100644
--- a/crates/bevy_pbr/src/light_probe/environment_map.rs
+++ b/crates/bevy_pbr/src/light_probe/environment_map.rs
@@ -65,7 +65,7 @@ use bevy_render::{
     texture::{FallbackImage, GpuImage, Image},
 };
 
-use std::num::NonZeroU32;
+use std::num::NonZero;
 use std::ops::Deref;
 
 use crate::{
@@ -217,7 +217,7 @@ pub(crate) fn get_bind_group_layout_entries(
         binding_types::texture_cube(TextureSampleType::Float { filterable: true });
     if binding_arrays_are_usable(render_device) {
         texture_cube_binding =
-            texture_cube_binding.count(NonZeroU32::new(MAX_VIEW_LIGHT_PROBES as _).unwrap());
+            texture_cube_binding.count(NonZero::<u32>::new(MAX_VIEW_LIGHT_PROBES as _).unwrap());
     }
 
     [
diff --git a/crates/bevy_pbr/src/light_probe/irradiance_volume.rs b/crates/bevy_pbr/src/light_probe/irradiance_volume.rs
index 618da04fa4490..58110e98afb29 100644
--- a/crates/bevy_pbr/src/light_probe/irradiance_volume.rs
+++ b/crates/bevy_pbr/src/light_probe/irradiance_volume.rs
@@ -142,7 +142,7 @@ use bevy_render::{
     renderer::RenderDevice,
     texture::{FallbackImage, GpuImage, Image},
 };
-use std::{num::NonZeroU32, ops::Deref};
+use std::{num::NonZero, ops::Deref};
 
 use bevy_asset::{AssetId, Handle};
 use bevy_reflect::Reflect;
@@ -306,7 +306,7 @@ pub(crate) fn get_bind_group_layout_entries(
         binding_types::texture_3d(TextureSampleType::Float { filterable: true });
     if binding_arrays_are_usable(render_device) {
         texture_3d_binding =
-            texture_3d_binding.count(NonZeroU32::new(MAX_VIEW_LIGHT_PROBES as _).unwrap());
+            texture_3d_binding.count(NonZero::<u32>::new(MAX_VIEW_LIGHT_PROBES as _).unwrap());
     }
 
     [
diff --git a/crates/bevy_pbr/src/material.rs b/crates/bevy_pbr/src/material.rs
index cee8129febe75..9da61b2ce9e64 100644
--- a/crates/bevy_pbr/src/material.rs
+++ b/crates/bevy_pbr/src/material.rs
@@ -35,7 +35,7 @@ use bevy_render::{
 use bevy_utils::tracing::error;
 use std::marker::PhantomData;
 use std::sync::atomic::{AtomicU32, Ordering};
-use std::{hash::Hash, num::NonZeroU32};
+use std::{hash::Hash, num::NonZero};
 
 use self::{irradiance_volume::IrradianceVolume, prelude::EnvironmentMapLight};
 
@@ -978,7 +978,7 @@ impl AtomicMaterialBindGroupId {
     /// See also:  [`AtomicU32::store`].
     pub fn set(&self, id: MaterialBindGroupId) {
         let id = if let Some(id) = id.0 {
-            NonZeroU32::from(id).get()
+            NonZero::<u32>::from(id).get()
         } else {
             0
         };
@@ -990,7 +990,9 @@ impl AtomicMaterialBindGroupId {
     ///
     /// See also:  [`AtomicU32::load`].
     pub fn get(&self) -> MaterialBindGroupId {
-        MaterialBindGroupId(NonZeroU32::new(self.0.load(Ordering::Relaxed)).map(BindGroupId::from))
+        MaterialBindGroupId(
+            NonZero::<u32>::new(self.0.load(Ordering::Relaxed)).map(BindGroupId::from),
+        )
     }
 }
 
diff --git a/crates/bevy_pbr/src/meshlet/persistent_buffer.rs b/crates/bevy_pbr/src/meshlet/persistent_buffer.rs
index 60e163a87446a..e10dad6ef0aed 100644
--- a/crates/bevy_pbr/src/meshlet/persistent_buffer.rs
+++ b/crates/bevy_pbr/src/meshlet/persistent_buffer.rs
@@ -6,7 +6,7 @@ use bevy_render::{
     renderer::{RenderDevice, RenderQueue},
 };
 use range_alloc::RangeAllocator;
-use std::{num::NonZeroU64, ops::Range};
+use std::{num::NonZero, ops::Range};
 
 /// Wrapper for a GPU buffer holding a large amount of data that persists across frames.
 pub struct PersistentGpuBuffer<T: PersistentGpuBufferable> {
@@ -66,7 +66,8 @@ impl<T: PersistentGpuBufferable> PersistentGpuBuffer<T> {
         let queue_count = self.write_queue.len();
 
         for (data, metadata, buffer_slice) in self.write_queue.drain(..) {
-            let buffer_slice_size = NonZeroU64::new(buffer_slice.end - buffer_slice.start).unwrap();
+            let buffer_slice_size =
+                NonZero::<u64>::new(buffer_slice.end - buffer_slice.start).unwrap();
             let mut buffer_view = render_queue
                 .write_buffer_with(&self.buffer, buffer_slice.start, buffer_slice_size)
                 .unwrap();
diff --git a/crates/bevy_pbr/src/render/gpu_preprocess.rs b/crates/bevy_pbr/src/render/gpu_preprocess.rs
index 67eca5df5e13d..38650d280891b 100644
--- a/crates/bevy_pbr/src/render/gpu_preprocess.rs
+++ b/crates/bevy_pbr/src/render/gpu_preprocess.rs
@@ -6,7 +6,7 @@
 //! [`MeshInputUniform`]s instead and use the GPU to calculate the remaining
 //! derived fields in [`MeshUniform`].
 
-use std::num::NonZeroU64;
+use std::num::NonZero;
 
 use bevy_app::{App, Plugin};
 use bevy_asset::{load_internal_asset, Handle};
@@ -408,7 +408,7 @@ pub fn prepare_preprocess_bind_groups(
         // Don't use `as_entire_binding()` here; the shader reads the array
         // length and the underlying buffer may be longer than the actual size
         // of the vector.
-        let index_buffer_size = NonZeroU64::try_from(
+        let index_buffer_size = NonZero::<u64>::try_from(
             index_buffer_vec.buffer.len() as u64 * u64::from(PreprocessWorkItem::min_size()),
         )
         .ok();
diff --git a/crates/bevy_pbr/src/render/mesh_view_bindings.rs b/crates/bevy_pbr/src/render/mesh_view_bindings.rs
index 36c5d7bfe044f..d0a506e9c3e42 100644
--- a/crates/bevy_pbr/src/render/mesh_view_bindings.rs
+++ b/crates/bevy_pbr/src/render/mesh_view_bindings.rs
@@ -1,4 +1,4 @@
-use std::{array, num::NonZeroU64, sync::Arc};
+use std::{array, num::NonZero, sync::Arc};
 
 use bevy_core_pipeline::{
     core_3d::ViewTransmissionTexture,
@@ -164,7 +164,7 @@ impl From<Option<&ViewPrepassTextures>> for MeshPipelineViewLayoutKey {
 fn buffer_layout(
     buffer_binding_type: BufferBindingType,
     has_dynamic_offset: bool,
-    min_binding_size: Option<NonZeroU64>,
+    min_binding_size: Option<NonZero<u64>>,
 ) -> BindGroupLayoutEntryBuilder {
     match buffer_binding_type {
         BufferBindingType::Uniform => uniform_buffer_sized(has_dynamic_offset, min_binding_size),
diff --git a/crates/bevy_ptr/src/lib.rs b/crates/bevy_ptr/src/lib.rs
index 24ea602296256..5b2186c44fbc5 100644
--- a/crates/bevy_ptr/src/lib.rs
+++ b/crates/bevy_ptr/src/lib.rs
@@ -12,7 +12,7 @@ use core::{
     fmt::{self, Formatter, Pointer},
     marker::PhantomData,
     mem::{align_of, ManuallyDrop},
-    num::NonZeroUsize,
+    num::NonZero,
     ptr::NonNull,
 };
 
@@ -535,10 +535,10 @@ impl<'a, T> From<&'a [T]> for ThinSlicePtr<'a, T> {
 
 /// Creates a dangling pointer with specified alignment.
 /// See [`NonNull::dangling`].
-pub fn dangling_with_align(align: NonZeroUsize) -> NonNull<u8> {
+pub fn dangling_with_align(align: NonZero<usize>) -> NonNull<u8> {
     debug_assert!(align.is_power_of_two(), "Alignment must be power of two.");
     // SAFETY: The pointer will not be null, since it was created
-    // from the address of a `NonZeroUsize`.
+    // from the address of a `NonZero<usize>`.
     unsafe { NonNull::new_unchecked(align.get() as *mut u8) }
 }
 
diff --git a/crates/bevy_reflect/src/impls/std.rs b/crates/bevy_reflect/src/impls/std.rs
index 6320090e6d4f6..c9f27a4cb34a4 100644
--- a/crates/bevy_reflect/src/impls/std.rs
+++ b/crates/bevy_reflect/src/impls/std.rs
@@ -2444,11 +2444,11 @@ mod tests {
 
     #[test]
     fn nonzero_usize_impl_reflect_from_reflect() {
-        let a: &dyn PartialReflect = &std::num::NonZeroUsize::new(42).unwrap();
-        let b: &dyn PartialReflect = &std::num::NonZeroUsize::new(42).unwrap();
+        let a: &dyn PartialReflect = &std::num::NonZero::<usize>::new(42).unwrap();
+        let b: &dyn PartialReflect = &std::num::NonZero::<usize>::new(42).unwrap();
         assert!(a.reflect_partial_eq(b).unwrap_or_default());
-        let forty_two: std::num::NonZeroUsize = FromReflect::from_reflect(a).unwrap();
-        assert_eq!(forty_two, std::num::NonZeroUsize::new(42).unwrap());
+        let forty_two: std::num::NonZero<usize> = FromReflect::from_reflect(a).unwrap();
+        assert_eq!(forty_two, std::num::NonZero::<usize>::new(42).unwrap());
     }
 
     #[test]
diff --git a/crates/bevy_render/src/render_resource/batched_uniform_buffer.rs b/crates/bevy_render/src/render_resource/batched_uniform_buffer.rs
index d92c7a3897e06..75a747cf46fe3 100644
--- a/crates/bevy_render/src/render_resource/batched_uniform_buffer.rs
+++ b/crates/bevy_render/src/render_resource/batched_uniform_buffer.rs
@@ -8,7 +8,7 @@ use encase::{
     ShaderType,
 };
 use nonmax::NonMaxU32;
-use std::{marker::PhantomData, num::NonZeroU64};
+use std::{marker::PhantomData, num::NonZero};
 use wgpu::{BindingResource, Limits};
 
 // 1MB else we will make really large arrays on macOS which reports very large
@@ -69,7 +69,7 @@ impl<T: GpuArrayBufferable> BatchedUniformBuffer<T> {
     }
 
     #[inline]
-    pub fn size(&self) -> NonZeroU64 {
+    pub fn size(&self) -> NonZero<u64> {
         self.temp.size()
     }
 
@@ -141,7 +141,7 @@ where
 
     const METADATA: Metadata<Self::ExtraMetadata> = T::METADATA;
 
-    fn size(&self) -> NonZeroU64 {
+    fn size(&self) -> NonZero<u64> {
         Self::METADATA.stride().mul(self.1.max(1) as u64).0
     }
 }
diff --git a/crates/bevy_render/src/render_resource/bind_group_layout_entries.rs b/crates/bevy_render/src/render_resource/bind_group_layout_entries.rs
index 05c5ee9c3e818..58d1d62a19889 100644
--- a/crates/bevy_render/src/render_resource/bind_group_layout_entries.rs
+++ b/crates/bevy_render/src/render_resource/bind_group_layout_entries.rs
@@ -1,5 +1,5 @@
 use bevy_utils::all_tuples_with_size;
-use std::num::NonZeroU32;
+use std::num::NonZero;
 use wgpu::{BindGroupLayoutEntry, BindingType, ShaderStages};
 
 /// Helper for constructing bind group layouts.
@@ -130,7 +130,7 @@ use wgpu::{BindGroupLayoutEntry, BindingType, ShaderStages};
 pub struct BindGroupLayoutEntryBuilder {
     ty: BindingType,
     visibility: Option<ShaderStages>,
-    count: Option<NonZeroU32>,
+    count: Option<NonZero<u32>>,
 }
 
 impl BindGroupLayoutEntryBuilder {
@@ -139,7 +139,7 @@ impl BindGroupLayoutEntryBuilder {
         self
     }
 
-    pub fn count(mut self, count: NonZeroU32) -> Self {
+    pub fn count(mut self, count: NonZero<u32>) -> Self {
         self.count = Some(count);
         self
     }
@@ -353,7 +353,7 @@ pub mod binding_types {
         BufferBindingType, SamplerBindingType, TextureSampleType, TextureViewDimension,
     };
     use encase::ShaderType;
-    use std::num::NonZeroU64;
+    use std::num::NonZero;
     use wgpu::{StorageTextureAccess, TextureFormat};
 
     use super::*;
@@ -364,7 +364,7 @@ pub mod binding_types {
 
     pub fn storage_buffer_sized(
         has_dynamic_offset: bool,
-        min_binding_size: Option<NonZeroU64>,
+        min_binding_size: Option<NonZero<u64>>,
     ) -> BindGroupLayoutEntryBuilder {
         BindingType::Buffer {
             ty: BufferBindingType::Storage { read_only: false },
@@ -382,7 +382,7 @@ pub mod binding_types {
 
     pub fn storage_buffer_read_only_sized(
         has_dynamic_offset: bool,
-        min_binding_size: Option<NonZeroU64>,
+        min_binding_size: Option<NonZero<u64>>,
     ) -> BindGroupLayoutEntryBuilder {
         BindingType::Buffer {
             ty: BufferBindingType::Storage { read_only: true },
@@ -398,7 +398,7 @@ pub mod binding_types {
 
     pub fn uniform_buffer_sized(
         has_dynamic_offset: bool,
-        min_binding_size: Option<NonZeroU64>,
+        min_binding_size: Option<NonZero<u64>>,
     ) -> BindGroupLayoutEntryBuilder {
         BindingType::Buffer {
             ty: BufferBindingType::Uniform,
diff --git a/crates/bevy_render/src/render_resource/resource_macros.rs b/crates/bevy_render/src/render_resource/resource_macros.rs
index 68896092ce016..e22c59b1c031f 100644
--- a/crates/bevy_render/src/render_resource/resource_macros.rs
+++ b/crates/bevy_render/src/render_resource/resource_macros.rs
@@ -149,7 +149,7 @@ macro_rules! render_resource_wrapper {
 macro_rules! define_atomic_id {
     ($atomic_id_type:ident) => {
         #[derive(Copy, Clone, Hash, Eq, PartialEq, PartialOrd, Ord, Debug)]
-        pub struct $atomic_id_type(core::num::NonZeroU32);
+        pub struct $atomic_id_type(core::num::NonZero<u32>);
 
         // We use new instead of default to indicate that each ID created will be unique.
         #[allow(clippy::new_without_default)]
@@ -160,7 +160,7 @@ macro_rules! define_atomic_id {
                 static COUNTER: AtomicU32 = AtomicU32::new(1);
 
                 let counter = COUNTER.fetch_add(1, Ordering::Relaxed);
-                Self(core::num::NonZeroU32::new(counter).unwrap_or_else(|| {
+                Self(core::num::NonZero::<u32>::new(counter).unwrap_or_else(|| {
                     panic!(
                         "The system ran out of unique `{}`s.",
                         stringify!($atomic_id_type)
@@ -169,14 +169,14 @@ macro_rules! define_atomic_id {
             }
         }
 
-        impl From<$atomic_id_type> for core::num::NonZeroU32 {
+        impl From<$atomic_id_type> for core::num::NonZero<u32> {
             fn from(value: $atomic_id_type) -> Self {
                 value.0
             }
         }
 
-        impl From<core::num::NonZeroU32> for $atomic_id_type {
-            fn from(value: core::num::NonZeroU32) -> Self {
+        impl From<core::num::NonZero<u32>> for $atomic_id_type {
+            fn from(value: core::num::NonZero<u32>) -> Self {
                 Self(value)
             }
         }
diff --git a/crates/bevy_render/src/render_resource/uniform_buffer.rs b/crates/bevy_render/src/render_resource/uniform_buffer.rs
index de4d84c94a06e..db51653f146ea 100644
--- a/crates/bevy_render/src/render_resource/uniform_buffer.rs
+++ b/crates/bevy_render/src/render_resource/uniform_buffer.rs
@@ -1,4 +1,4 @@
-use std::{marker::PhantomData, num::NonZeroU64};
+use std::{marker::PhantomData, num::NonZero};
 
 use crate::{
     render_resource::Buffer,
@@ -309,7 +309,7 @@ impl<T: ShaderType + WriteInto> DynamicUniformBuffer<T> {
 
         if let Some(buffer) = self.buffer.as_deref() {
             let buffer_view = queue
-                .write_buffer_with(buffer, 0, NonZeroU64::new(buffer.size())?)
+                .write_buffer_with(buffer, 0, NonZero::<u64>::new(buffer.size())?)
                 .unwrap();
             Some(DynamicUniformBufferWriter {
                 buffer: encase::DynamicUniformBuffer::new_with_alignment(
diff --git a/crates/bevy_render/src/view/window/mod.rs b/crates/bevy_render/src/view/window/mod.rs
index 816d8c4e8dfd7..d9d30d440dcf5 100644
--- a/crates/bevy_render/src/view/window/mod.rs
+++ b/crates/bevy_render/src/view/window/mod.rs
@@ -13,7 +13,7 @@ use bevy_window::{
 };
 use bevy_winit::CustomCursorCache;
 use std::{
-    num::NonZeroU32,
+    num::NonZero,
     ops::{Deref, DerefMut},
 };
 use wgpu::{
@@ -63,7 +63,7 @@ pub struct ExtractedWindow {
     pub physical_width: u32,
     pub physical_height: u32,
     pub present_mode: PresentMode,
-    pub desired_maximum_frame_latency: Option<NonZeroU32>,
+    pub desired_maximum_frame_latency: Option<NonZero<u32>>,
     /// Note: this will not always be the swap chain texture view. When taking a screenshot,
     /// this will point to an alternative texture instead to allow for copying the render result
     /// to CPU memory.
@@ -395,7 +395,7 @@ pub fn create_surfaces(
                     },
                     desired_maximum_frame_latency: window
                         .desired_maximum_frame_latency
-                        .map(NonZeroU32::get)
+                        .map(NonZero::<u32>::get)
                         .unwrap_or(DEFAULT_DESIRED_MAXIMUM_FRAME_LATENCY),
                     alpha_mode: match window.alpha_mode {
                         CompositeAlphaMode::Auto => wgpu::CompositeAlphaMode::Auto,
diff --git a/crates/bevy_tasks/src/lib.rs b/crates/bevy_tasks/src/lib.rs
index 3eb33c6603e70..1fe4f9511d997 100644
--- a/crates/bevy_tasks/src/lib.rs
+++ b/crates/bevy_tasks/src/lib.rs
@@ -55,7 +55,7 @@ pub mod prelude {
     };
 }
 
-use std::num::NonZeroUsize;
+use std::num::NonZero;
 
 /// Gets the logical CPU core count available to the current process.
 ///
@@ -65,6 +65,6 @@ use std::num::NonZeroUsize;
 /// This will always return at least 1.
 pub fn available_parallelism() -> usize {
     std::thread::available_parallelism()
-        .map(NonZeroUsize::get)
+        .map(NonZero::<usize>::get)
         .unwrap_or(1)
 }
diff --git a/crates/bevy_ui/src/ui_node.rs b/crates/bevy_ui/src/ui_node.rs
index fb285823b3798..3a46e952b32a7 100644
--- a/crates/bevy_ui/src/ui_node.rs
+++ b/crates/bevy_ui/src/ui_node.rs
@@ -12,7 +12,7 @@ use bevy_transform::prelude::GlobalTransform;
 use bevy_utils::warn_once;
 use bevy_window::{PrimaryWindow, WindowRef};
 use smallvec::SmallVec;
-use std::num::{NonZeroI16, NonZeroU16};
+use std::num::NonZero;
 use thiserror::Error;
 
 /// Base component for a UI node, which also provides the computed size of the node.
@@ -1481,15 +1481,15 @@ pub struct GridPlacement {
     /// Lines are 1-indexed.
     /// Negative indexes count backwards from the end of the grid.
     /// Zero is not a valid index.
-    pub(crate) start: Option<NonZeroI16>,
+    pub(crate) start: Option<NonZero<i16>>,
     /// How many grid tracks the item should span.
     /// Defaults to 1.
-    pub(crate) span: Option<NonZeroU16>,
+    pub(crate) span: Option<NonZero<u16>>,
     /// The grid line at which the item should end.
     /// Lines are 1-indexed.
     /// Negative indexes count backwards from the end of the grid.
     /// Zero is not a valid index.
-    pub(crate) end: Option<NonZeroI16>,
+    pub(crate) end: Option<NonZero<i16>>,
 }
 
 impl GridPlacement {
@@ -1497,7 +1497,7 @@ impl GridPlacement {
     pub const DEFAULT: Self = Self {
         start: None,
         // SAFETY: This is trivially safe as 1 is non-zero.
-        span: Some(unsafe { NonZeroU16::new_unchecked(1) }),
+        span: Some(unsafe { NonZero::<u16>::new_unchecked(1) }),
         end: None,
     };
 
@@ -1614,17 +1614,17 @@ impl GridPlacement {
 
     /// Returns the grid line at which the item should start, or `None` if not set.
     pub fn get_start(self) -> Option<i16> {
-        self.start.map(NonZeroI16::get)
+        self.start.map(NonZero::<i16>::get)
     }
 
     /// Returns the grid line at which the item should end, or `None` if not set.
     pub fn get_end(self) -> Option<i16> {
-        self.end.map(NonZeroI16::get)
+        self.end.map(NonZero::<i16>::get)
     }
 
     /// Returns span for this grid item, or `None` if not set.
     pub fn get_span(self) -> Option<u16> {
-        self.span.map(NonZeroU16::get)
+        self.span.map(NonZero::<u16>::get)
     }
 }
 
@@ -1634,17 +1634,17 @@ impl Default for GridPlacement {
     }
 }
 
-/// Convert an `i16` to `NonZeroI16`, fails on `0` and returns the `InvalidZeroIndex` error.
-fn try_into_grid_index(index: i16) -> Result<Option<NonZeroI16>, GridPlacementError> {
+/// Convert an `i16` to `NonZero<i16>`, fails on `0` and returns the `InvalidZeroIndex` error.
+fn try_into_grid_index(index: i16) -> Result<Option<NonZero<i16>>, GridPlacementError> {
     Ok(Some(
-        NonZeroI16::new(index).ok_or(GridPlacementError::InvalidZeroIndex)?,
+        NonZero::<i16>::new(index).ok_or(GridPlacementError::InvalidZeroIndex)?,
     ))
 }
 
-/// Convert a `u16` to `NonZeroU16`, fails on `0` and returns the `InvalidZeroSpan` error.
-fn try_into_grid_span(span: u16) -> Result<Option<NonZeroU16>, GridPlacementError> {
+/// Convert a `u16` to `NonZero<u16>`, fails on `0` and returns the `InvalidZeroSpan` error.
+fn try_into_grid_span(span: u16) -> Result<Option<NonZero<u16>>, GridPlacementError> {
     Ok(Some(
-        NonZeroU16::new(span).ok_or(GridPlacementError::InvalidZeroSpan)?,
+        NonZero::<u16>::new(span).ok_or(GridPlacementError::InvalidZeroSpan)?,
     ))
 }
 
diff --git a/crates/bevy_window/src/window.rs b/crates/bevy_window/src/window.rs
index dfaa0ecdae36b..eb8ca610fc9af 100644
--- a/crates/bevy_window/src/window.rs
+++ b/crates/bevy_window/src/window.rs
@@ -1,4 +1,4 @@
-use std::num::NonZeroU32;
+use std::num::NonZero;
 
 use bevy_ecs::{
     entity::{Entity, EntityMapper, MapEntities},
@@ -279,7 +279,7 @@ pub struct Window {
     ///
     /// [`wgpu::SurfaceConfiguration::desired_maximum_frame_latency`]:
     /// https://docs.rs/wgpu/latest/wgpu/type.SurfaceConfiguration.html#structfield.desired_maximum_frame_latency
-    pub desired_maximum_frame_latency: Option<NonZeroU32>,
+    pub desired_maximum_frame_latency: Option<NonZero<u32>>,
     /// Sets whether this window recognizes [`PinchGesture`](https://docs.rs/bevy/latest/bevy/input/gestures/struct.PinchGesture.html)
     ///
     /// ## Platform-specific
diff --git a/examples/app/headless_renderer.rs b/examples/app/headless_renderer.rs
index 2a882063d9abf..f6fa5c347257d 100644
--- a/examples/app/headless_renderer.rs
+++ b/examples/app/headless_renderer.rs
@@ -381,7 +381,7 @@ impl render_graph::Node for ImageCopyDriver {
                     layout: ImageDataLayout {
                         offset: 0,
                         bytes_per_row: Some(
-                            std::num::NonZeroU32::new(padded_bytes_per_row as u32)
+                            std::num::NonZero::<u32>::new(padded_bytes_per_row as u32)
                                 .unwrap()
                                 .into(),
                         ),
diff --git a/examples/shader/texture_binding_array.rs b/examples/shader/texture_binding_array.rs
index 53e774df0485e..24a8a847a1c19 100644
--- a/examples/shader/texture_binding_array.rs
+++ b/examples/shader/texture_binding_array.rs
@@ -17,7 +17,7 @@ use bevy::{
         RenderApp,
     },
 };
-use std::{num::NonZeroU32, process::exit};
+use std::{num::NonZero, process::exit};
 
 /// This example uses a shader source file from the assets subdirectory
 const SHADER_ASSET_PATH: &str = "shaders/texture_binding_array.wgsl";
@@ -166,7 +166,7 @@ impl AsBindGroup for BindlessMaterial {
                 (
                     0,
                     texture_2d(TextureSampleType::Float { filterable: true })
-                        .count(NonZeroU32::new(MAX_TEXTURE_COUNT as u32).unwrap()),
+                        .count(NonZero::<u32>::new(MAX_TEXTURE_COUNT as u32).unwrap()),
                 ),
                 // Sampler
                 //
@@ -177,7 +177,7 @@ impl AsBindGroup for BindlessMaterial {
                 //
                 // ```
                 // sampler(SamplerBindingType::Filtering)
-                //     .count(NonZeroU32::new(MAX_TEXTURE_COUNT as u32).unwrap()),
+                //     .count(NonZero::<u32>::new(MAX_TEXTURE_COUNT as u32).unwrap()),
                 // ```
                 //
                 // One may need to pay attention to the limit of sampler binding