diff --git a/crates/accelerate/src/convert_2q_block_matrix.rs b/crates/accelerate/src/convert_2q_block_matrix.rs
index e311c129b11b..de72287293a7 100644
--- a/crates/accelerate/src/convert_2q_block_matrix.rs
+++ b/crates/accelerate/src/convert_2q_block_matrix.rs
@@ -74,7 +74,6 @@ pub fn change_basis(matrix: ArrayView2<Complex64>) -> Array2<Complex64> {
     trans_matrix
 }
 
-#[pymodule]
 pub fn convert_2q_block_matrix(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(blocks_to_matrix))?;
     Ok(())
diff --git a/crates/accelerate/src/dense_layout.rs b/crates/accelerate/src/dense_layout.rs
index 7cb54140761e..488f7d023cb4 100644
--- a/crates/accelerate/src/dense_layout.rs
+++ b/crates/accelerate/src/dense_layout.rs
@@ -223,7 +223,6 @@ pub fn best_subset(
     ))
 }
 
-#[pymodule]
 pub fn dense_layout(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(best_subset))?;
     Ok(())
diff --git a/crates/accelerate/src/error_map.rs b/crates/accelerate/src/error_map.rs
index 8dbd2e4290a5..b61733ae1512 100644
--- a/crates/accelerate/src/error_map.rs
+++ b/crates/accelerate/src/error_map.rs
@@ -111,7 +111,6 @@ impl ErrorMap {
     }
 }
 
-#[pymodule]
 pub fn error_map(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_class::<ErrorMap>()?;
     Ok(())
diff --git a/crates/accelerate/src/euler_one_qubit_decomposer.rs b/crates/accelerate/src/euler_one_qubit_decomposer.rs
index 1fd5fd7834ff..399a0d5f5fa9 100644
--- a/crates/accelerate/src/euler_one_qubit_decomposer.rs
+++ b/crates/accelerate/src/euler_one_qubit_decomposer.rs
@@ -883,7 +883,6 @@ pub fn params_zxz(unitary: PyReadonlyArray2<Complex64>) -> [f64; 4] {
     params_zxz_inner(mat)
 }
 
-#[pymodule]
 pub fn euler_one_qubit_decomposer(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(params_zyz))?;
     m.add_wrapped(wrap_pyfunction!(params_xyx))?;
diff --git a/crates/accelerate/src/isometry.rs b/crates/accelerate/src/isometry.rs
index a4e83358a7da..3ed3127eb562 100644
--- a/crates/accelerate/src/isometry.rs
+++ b/crates/accelerate/src/isometry.rs
@@ -348,7 +348,6 @@ fn b(k: usize, s: usize) -> usize {
     k - (a(k, s) * 2_usize.pow(s as u32))
 }
 
-#[pymodule]
 pub fn isometry(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(diag_is_identity_up_to_global_phase))?;
     m.add_wrapped(wrap_pyfunction!(find_squs_for_disentangling))?;
diff --git a/crates/accelerate/src/nlayout.rs b/crates/accelerate/src/nlayout.rs
index 1a0b73b25fed..b6b7be5613c3 100644
--- a/crates/accelerate/src/nlayout.rs
+++ b/crates/accelerate/src/nlayout.rs
@@ -216,7 +216,6 @@ impl NLayout {
     }
 }
 
-#[pymodule]
 pub fn nlayout(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_class::<NLayout>()?;
     Ok(())
diff --git a/crates/accelerate/src/optimize_1q_gates.rs b/crates/accelerate/src/optimize_1q_gates.rs
index a683604c19e0..64924a339136 100644
--- a/crates/accelerate/src/optimize_1q_gates.rs
+++ b/crates/accelerate/src/optimize_1q_gates.rs
@@ -90,7 +90,6 @@ pub fn compose_u3_rust(
     out_angles
 }
 
-#[pymodule]
 pub fn optimize_1q_gates(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(compose_u3_rust))?;
     Ok(())
diff --git a/crates/accelerate/src/pauli_exp_val.rs b/crates/accelerate/src/pauli_exp_val.rs
index 29f741f6cf46..66e07c4621a4 100644
--- a/crates/accelerate/src/pauli_exp_val.rs
+++ b/crates/accelerate/src/pauli_exp_val.rs
@@ -192,7 +192,6 @@ pub fn density_expval_pauli_with_x(
     }
 }
 
-#[pymodule]
 pub fn pauli_expval(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(expval_pauli_no_x))?;
     m.add_wrapped(wrap_pyfunction!(expval_pauli_with_x))?;
diff --git a/crates/accelerate/src/results/mod.rs b/crates/accelerate/src/results/mod.rs
index 36282a749e66..6d5a79363977 100644
--- a/crates/accelerate/src/results/mod.rs
+++ b/crates/accelerate/src/results/mod.rs
@@ -16,7 +16,6 @@ pub mod marginalization;
 use pyo3::prelude::*;
 use pyo3::wrap_pyfunction;
 
-#[pymodule]
 pub fn results(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(marginalization::marginal_counts))?;
     m.add_wrapped(wrap_pyfunction!(marginalization::marginal_distribution))?;
diff --git a/crates/accelerate/src/sabre/mod.rs b/crates/accelerate/src/sabre/mod.rs
index 3eb8ebb3a219..6305fb8f3bf7 100644
--- a/crates/accelerate/src/sabre/mod.rs
+++ b/crates/accelerate/src/sabre/mod.rs
@@ -113,7 +113,6 @@ impl BlockResult {
     }
 }
 
-#[pymodule]
 pub fn sabre(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(route::sabre_routing))?;
     m.add_wrapped(wrap_pyfunction!(layout::sabre_layout_and_routing))?;
diff --git a/crates/accelerate/src/sampled_exp_val.rs b/crates/accelerate/src/sampled_exp_val.rs
index b51ca3c98f0e..28cb4981df4a 100644
--- a/crates/accelerate/src/sampled_exp_val.rs
+++ b/crates/accelerate/src/sampled_exp_val.rs
@@ -86,7 +86,6 @@ pub fn sampled_expval_complex(
     Ok(out.re)
 }
 
-#[pymodule]
 pub fn sampled_exp_val(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(sampled_expval_float))?;
     m.add_wrapped(wrap_pyfunction!(sampled_expval_complex))?;
diff --git a/crates/accelerate/src/sparse_pauli_op.rs b/crates/accelerate/src/sparse_pauli_op.rs
index 5d6a82df7940..e17d2faca713 100644
--- a/crates/accelerate/src/sparse_pauli_op.rs
+++ b/crates/accelerate/src/sparse_pauli_op.rs
@@ -816,7 +816,6 @@ impl_to_matrix_sparse!(
     u64
 );
 
-#[pymodule]
 pub fn sparse_pauli_op(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(unordered_unique))?;
     m.add_wrapped(wrap_pyfunction!(decompose_dense))?;
diff --git a/crates/accelerate/src/stochastic_swap.rs b/crates/accelerate/src/stochastic_swap.rs
index bc13325d8d96..7b18500240eb 100644
--- a/crates/accelerate/src/stochastic_swap.rs
+++ b/crates/accelerate/src/stochastic_swap.rs
@@ -335,7 +335,6 @@ pub fn swap_trials(
     Ok((best_edges, best_layout, best_depth))
 }
 
-#[pymodule]
 pub fn stochastic_swap(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(swap_trials))?;
     m.add_class::<EdgeCollection>()?;
diff --git a/crates/accelerate/src/two_qubit_decompose.rs b/crates/accelerate/src/two_qubit_decompose.rs
index 5e833bd86fda..3750f6b006b0 100644
--- a/crates/accelerate/src/two_qubit_decompose.rs
+++ b/crates/accelerate/src/two_qubit_decompose.rs
@@ -2042,7 +2042,6 @@ impl TwoQubitBasisDecomposer {
     }
 }
 
-#[pymodule]
 pub fn two_qubit_decompose(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(_num_basis_gates))?;
     m.add_class::<TwoQubitGateSequence>()?;
diff --git a/crates/accelerate/src/uc_gate.rs b/crates/accelerate/src/uc_gate.rs
index 3a5f74a6f0b1..917361389d63 100644
--- a/crates/accelerate/src/uc_gate.rs
+++ b/crates/accelerate/src/uc_gate.rs
@@ -156,7 +156,6 @@ pub fn dec_ucg_help(
     )
 }
 
-#[pymodule]
 pub fn uc_gate(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(dec_ucg_help))?;
     Ok(())
diff --git a/crates/accelerate/src/utils.rs b/crates/accelerate/src/utils.rs
index 6df00f7f8b76..598256192f83 100644
--- a/crates/accelerate/src/utils.rs
+++ b/crates/accelerate/src/utils.rs
@@ -41,7 +41,6 @@ pub fn eigenvalues(py: Python, unitary: PyReadonlyArray2<Complex<f64>>) -> PyObj
         .into()
 }
 
-#[pymodule]
 pub fn utils(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(eigenvalues))?;
     Ok(())
diff --git a/crates/accelerate/src/vf2_layout.rs b/crates/accelerate/src/vf2_layout.rs
index e84b2bfbfb8c..476197397aff 100644
--- a/crates/accelerate/src/vf2_layout.rs
+++ b/crates/accelerate/src/vf2_layout.rs
@@ -106,7 +106,6 @@ pub fn score_layout(
     Ok(1. - fidelity)
 }
 
-#[pymodule]
 pub fn vf2_layout(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_wrapped(wrap_pyfunction!(score_layout))?;
     m.add_class::<EdgeList>()?;
diff --git a/crates/circuit/src/lib.rs b/crates/circuit/src/lib.rs
index cd560bad7387..e74c19d675c3 100644
--- a/crates/circuit/src/lib.rs
+++ b/crates/circuit/src/lib.rs
@@ -27,8 +27,7 @@ pub enum SliceOrInt<'a> {
     Slice(Bound<'a, PySlice>),
 }
 
-#[pymodule]
-pub fn circuit(m: Bound<PyModule>) -> PyResult<()> {
+pub fn circuit(m: &Bound<PyModule>) -> PyResult<()> {
     m.add_class::<circuit_data::CircuitData>()?;
     m.add_class::<circuit_instruction::CircuitInstruction>()?;
     Ok(())
diff --git a/crates/pyext/src/lib.rs b/crates/pyext/src/lib.rs
index a21b1307a88f..b7874d740906 100644
--- a/crates/pyext/src/lib.rs
+++ b/crates/pyext/src/lib.rs
@@ -11,7 +11,6 @@
 // that they have been altered from the originals.
 
 use pyo3::prelude::*;
-use pyo3::wrap_pymodule;
 
 use qiskit_accelerate::{
     convert_2q_block_matrix::convert_2q_block_matrix, dense_layout::dense_layout,
@@ -23,27 +22,38 @@ use qiskit_accelerate::{
     vf2_layout::vf2_layout,
 };
 
+#[inline(always)]
+#[doc(hidden)]
+fn add_submodule<F>(m: &Bound<PyModule>, constructor: F, name: &str) -> PyResult<()>
+where
+    F: FnOnce(&Bound<PyModule>) -> PyResult<()>,
+{
+    let new_mod = PyModule::new_bound(m.py(), name)?;
+    constructor(&new_mod)?;
+    m.add_submodule(&new_mod)
+}
+
 #[pymodule]
 fn _accelerate(m: &Bound<PyModule>) -> PyResult<()> {
-    m.add_wrapped(wrap_pymodule!(qiskit_circuit::circuit))?;
-    m.add_wrapped(wrap_pymodule!(qiskit_qasm2::qasm2))?;
-    m.add_wrapped(wrap_pymodule!(qiskit_qasm3::qasm3))?;
-    m.add_wrapped(wrap_pymodule!(convert_2q_block_matrix))?;
-    m.add_wrapped(wrap_pymodule!(dense_layout))?;
-    m.add_wrapped(wrap_pymodule!(error_map))?;
-    m.add_wrapped(wrap_pymodule!(euler_one_qubit_decomposer))?;
-    m.add_wrapped(wrap_pymodule!(isometry))?;
-    m.add_wrapped(wrap_pymodule!(nlayout))?;
-    m.add_wrapped(wrap_pymodule!(optimize_1q_gates))?;
-    m.add_wrapped(wrap_pymodule!(pauli_expval))?;
-    m.add_wrapped(wrap_pymodule!(results))?;
-    m.add_wrapped(wrap_pymodule!(sabre))?;
-    m.add_wrapped(wrap_pymodule!(sampled_exp_val))?;
-    m.add_wrapped(wrap_pymodule!(sparse_pauli_op))?;
-    m.add_wrapped(wrap_pymodule!(stochastic_swap))?;
-    m.add_wrapped(wrap_pymodule!(two_qubit_decompose))?;
-    m.add_wrapped(wrap_pymodule!(uc_gate))?;
-    m.add_wrapped(wrap_pymodule!(utils))?;
-    m.add_wrapped(wrap_pymodule!(vf2_layout))?;
+    add_submodule(m, qiskit_circuit::circuit, "circuit")?;
+    add_submodule(m, qiskit_qasm2::qasm2, "qasm2")?;
+    add_submodule(m, qiskit_qasm3::qasm3, "qasm3")?;
+    add_submodule(m, convert_2q_block_matrix, "convert_2q_block_matrix")?;
+    add_submodule(m, dense_layout, "dense_layout")?;
+    add_submodule(m, error_map, "error_map")?;
+    add_submodule(m, euler_one_qubit_decomposer, "euler_one_qubit_decomposer")?;
+    add_submodule(m, isometry, "isometry")?;
+    add_submodule(m, nlayout, "nlayout")?;
+    add_submodule(m, optimize_1q_gates, "optimize_1q_gates")?;
+    add_submodule(m, pauli_expval, "pauli_expval")?;
+    add_submodule(m, results, "results")?;
+    add_submodule(m, sabre, "sabre")?;
+    add_submodule(m, sampled_exp_val, "sampled_exp_val")?;
+    add_submodule(m, sparse_pauli_op, "sparse_pauli_op")?;
+    add_submodule(m, stochastic_swap, "stochastic_swap")?;
+    add_submodule(m, two_qubit_decompose, "two_qubit_decompose")?;
+    add_submodule(m, uc_gate, "uc_gate")?;
+    add_submodule(m, utils, "utils")?;
+    add_submodule(m, vf2_layout, "vf2_layout")?;
     Ok(())
 }
diff --git a/crates/qasm2/src/lib.rs b/crates/qasm2/src/lib.rs
index 7129b17f7fdf..beac72674e99 100644
--- a/crates/qasm2/src/lib.rs
+++ b/crates/qasm2/src/lib.rs
@@ -123,7 +123,6 @@ fn bytecode_from_file(
 /// An interface to the Rust components of the parser stack, and the types it uses to represent the
 /// output.  The principal entry points for Python are :func:`bytecode_from_string` and
 /// :func:`bytecode_from_file`, which produce iterables of :class:`Bytecode` objects.
-#[pymodule]
 pub fn qasm2(module: &Bound<PyModule>) -> PyResult<()> {
     module.add_class::<bytecode::OpCode>()?;
     module.add_class::<bytecode::UnaryOpCode>()?;
diff --git a/crates/qasm3/src/lib.rs b/crates/qasm3/src/lib.rs
index 9a651a5c12e2..bca862406f39 100644
--- a/crates/qasm3/src/lib.rs
+++ b/crates/qasm3/src/lib.rs
@@ -153,7 +153,6 @@ pub fn load(
 
 /// Internal module supplying the OpenQASM 3 import capabilities.  The entries in it should largely
 /// be re-exposed directly to public Python space.
-#[pymodule]
 pub fn qasm3(module: &Bound<PyModule>) -> PyResult<()> {
     module.add_function(wrap_pyfunction!(loads, module)?)?;
     module.add_function(wrap_pyfunction!(load, module)?)?;