diff --git a/crates/sol-types/src/abi/encoder.rs b/crates/sol-types/src/abi/encoder.rs
index c802f5798..23fd77d8d 100644
--- a/crates/sol-types/src/abi/encoder.rs
+++ b/crates/sol-types/src/abi/encoder.rs
@@ -43,12 +43,6 @@ impl Encoder {
         }
     }
 
-    /// Return a reference to the encoded words.
-    #[inline]
-    pub fn words(&self) -> &[Word] {
-        &self.buf
-    }
-
     /// Finish the encoding process, returning the encoded words.
     ///
     /// Use `into_bytes` instead to flatten the words into bytes.
@@ -59,18 +53,16 @@ impl Encoder {
         self.buf
     }
 
-    /// Return a reference to the encoded bytes.
-    #[inline]
-    pub fn bytes(&self) -> &[u8] {
-        // SAFETY: `#[repr(transparent)] FixedBytes<N>([u8; N])`
-        unsafe { &*(self.words() as *const [Word] as *const [[u8; 32]]) }.as_flattened()
-    }
-
     /// Finish the encoding process, returning the encoded bytes.
     #[inline]
     pub fn into_bytes(self) -> Vec<u8> {
+        // TODO: remove once `Vec::into_flattened` is stabilized.
+        // unsafe { mem::transmute::<Vec<_>, Vec<[u8; 32]>>(self.buf) }.into_flattened()
+
         // SAFETY: `#[repr(transparent)] FixedBytes<N>([u8; N])`
-        unsafe { mem::transmute::<Vec<Word>, Vec<[u8; 32]>>(self.finish()) }.into_flattened()
+        crate::impl_core::into_flattened::<u8, 32>(unsafe {
+            mem::transmute::<Vec<Word>, Vec<[u8; 32]>>(self.buf)
+        })
     }
 
     /// Determine the current suffix offset.
diff --git a/crates/sol-types/src/impl_core.rs b/crates/sol-types/src/impl_core.rs
index 4f6e114c2..505987d10 100644
--- a/crates/sol-types/src/impl_core.rs
+++ b/crates/sol-types/src/impl_core.rs
@@ -1,7 +1,16 @@
 //! Modified implementations of unstable libcore functions.
 
+use alloc::vec::Vec;
 use core::mem::{self, MaybeUninit};
 
+trait Ext {
+    const IS_ZST: bool;
+}
+
+impl<T> Ext for T {
+    const IS_ZST: bool = mem::size_of::<Self>() == 0;
+}
+
 /// [`core::array::try_from_fn`]
 #[inline]
 pub(crate) fn try_from_fn<F, T, E, const N: usize>(mut cb: F) -> Result<[T; N], E>
@@ -79,3 +88,34 @@ pub(crate) unsafe fn array_assume_init<T, const N: usize>(array: [MaybeUninit<T>
 unsafe fn transpose<T, const N: usize>(array: [MaybeUninit<T>; N]) -> MaybeUninit<[T; N]> {
     mem::transmute_copy::<[MaybeUninit<T>; N], MaybeUninit<[T; N]>>(&mem::ManuallyDrop::new(&array))
 }
+
+// TODO(MSRV-1.80): remove
+/// [`Vec::into_flattened`].
+#[inline]
+pub(crate) fn into_flattened<T, const N: usize>(vec: Vec<[T; N]>) -> Vec<T> {
+    let (ptr, len, cap) = into_raw_parts(vec);
+    let (new_len, new_cap) = if T::IS_ZST {
+        (len.checked_mul(N).expect("vec len overflow"), usize::MAX)
+    } else {
+        // SAFETY:
+        // - `cap * N` cannot overflow because the allocation is already in
+        // the address space.
+        // - Each `[T; N]` has `N` valid elements, so there are `len * N`
+        // valid elements in the allocation.
+        unsafe { (len.checked_mul(N).unwrap_unchecked(), cap.checked_mul(N).unwrap_unchecked()) }
+    };
+    // SAFETY:
+    // - `ptr` was allocated by `self`
+    // - `ptr` is well-aligned because `[T; N]` has the same alignment as `T`.
+    // - `new_cap` refers to the same sized allocation as `cap` because
+    // `new_cap * size_of::<T>()` == `cap * size_of::<[T; N]>()`
+    // - `len` <= `cap`, so `len * N` <= `cap * N`.
+    unsafe { Vec::from_raw_parts(ptr.cast(), new_len, new_cap) }
+}
+
+/// [`Vec::into_raw_parts`]
+#[inline(always)]
+fn into_raw_parts<T>(vec: Vec<T>) -> (*mut T, usize, usize) {
+    let mut me = mem::ManuallyDrop::new(vec);
+    (me.as_mut_ptr(), me.len(), me.capacity())
+}