diff --git a/README.md b/README.md
index 2d807c7..8c52419 100644
--- a/README.md
+++ b/README.md
@@ -68,7 +68,7 @@ pixi run plot     # Run benchmarks and plot their results in data/
 from qlabs.base import StateVector, Hadamard, SWAP, NOT, PauliY, PauliZ
 from qlabs.abstractions import GateCircuit, StateVectorSimulator, ShowAfterEachGate
 
-num_qubits = 3
+alias num_qubits = 3
 qc = GateCircuit(num_qubits)
 
 qc.apply_gates(
@@ -92,11 +92,31 @@ print("Quantum circuit created:\n", qc)  # Visualization not fully implemented
 qsimu = StateVectorSimulator(
     qc,
     initial_state=StateVector.from_bitstring("0" * num_qubits),
-    use_gpu_if_available=True,  # GPU support not fully implemented
+    use_gpu_if_available=False,
     verbose=True,
     verbose_step_size=ShowAfterEachGate,  # Options: ShowAfterEachGate, ShowOnlyEnd
 )
 
+# GPU usage
+# The StateVectorSimulator currently does not support GPU with control gates fully.
+# The lower-level function `qubit_wise_multiply()` is correctly implemented for GPU.
+# qsimu = StateVectorSimulator[
+#     gpu_num_qubits=num_qubits,
+#     gpu_gate_ordered_set= [Hadamard, NOT, PauliY, PauliZ],
+#     gpu_control_gate_count=3,  # Planned to be automated
+#     gpu_control_bits_list= [
+#         [[2, 0]],
+#         [[1, 1]],
+#         [[0, 1]],
+#     ],  # Planned to be automated
+# ](
+#     qc,
+#     initial_state=StateVector.from_bitstring("0" * num_qubits),
+#     use_gpu_if_available=True,
+#     verbose=True,
+#     verbose_step_size=ShowOnlyEnd,
+# )
+
 final_state = qsimu.run()
 
 print("Final quantum state:\n", final_state)
diff --git a/examples/circuit_level.mojo b/examples/circuit_level.mojo
index 13192f3..843f12b 100644
--- a/examples/circuit_level.mojo
+++ b/examples/circuit_level.mojo
@@ -56,26 +56,27 @@ fn simulate_figure1_circuit_abstract() -> None:
     # Create the initial state |000⟩
     initial_state: StateVector = StateVector.from_bitstring("000")
 
+    # Simulating using CPU
     qsimu = StateVectorSimulator(
         qc,
         initial_state=initial_state,
-        use_gpu_if_available=True,
+        use_gpu_if_available=False,
         verbose=True,
-        verbose_step_size=ShowOnlyEnd,  # ShowAfterEachGate, ShowAfterEachLayer
+        verbose_step_size=ShowOnlyEnd,  # or ShowAfterEachGate, ShowAfterEachLayer
     )
 
-    # # For GPU (NOTE: Doesn't work currently as GPU doesn't fully support control bits)
+    # # Simulating using GPU
     # qsimu = StateVectorSimulator[
     #     gpu_num_qubits=num_qubits,
     #     gpu_gate_ordered_set= [Hadamard, PauliZ, NOT],
     #     gpu_control_gate_count=2,
-    #     gpu_control_bits_list= [[[1, 1], [1, 1]]],
+    #     gpu_control_bits_list= [[[1, 1]], [[1, 1]]],
     # ](
     #     qc,
     #     initial_state=initial_state,
     #     use_gpu_if_available=True,
     #     verbose=True,
-    #     verbose_step_size=ShowOnlyEnd,  # ShowAfterEachGate, ShowAfterEachLayer
+    #     verbose_step_size=ShowOnlyEnd,
     # )
 
     state = qsimu.run()
@@ -122,7 +123,9 @@ fn simulate_figure4_circuit_abstract() -> None:
     print("Final quantum state:\n", final_state)
 
 
-fn simulate_random_circuit[number_qubits: Int](number_layers: Int) -> None:
+fn simulate_random_circuit[
+    number_qubits: Int, use_gpu: Bool = True
+](number_layers: Int) -> None:
     """Simulates a random quantum circuit with the specified number of qubits and layers.
 
     Parameters:
@@ -167,7 +170,7 @@ fn simulate_random_circuit[number_qubits: Int](number_layers: Int) -> None:
     ](
         qc,
         initial_state=quantum_state,
-        use_gpu_if_available=True,
+        use_gpu_if_available=use_gpu,
         verbose=False,
     )
 
diff --git a/examples/main.mojo b/examples/main.mojo
index 551f098..d49e361 100644
--- a/examples/main.mojo
+++ b/examples/main.mojo
@@ -46,7 +46,9 @@ def main():
 
     print("=========================================")
     print("circuit_level: simulate_random_circuit()")
-    simulate_random_circuit[number_qubits=20](number_layers=5)
+    simulate_random_circuit[number_qubits=20, use_gpu=True](
+        number_layers=5,
+    )
 
     print("=========================================")
     print("circuit_level: simulate_figure4_circuit_abstract()")
diff --git a/img/benchmark_H100.png b/img/benchmark_H100.png
index 0dfea41..8ce2564 100644
Binary files a/img/benchmark_H100.png and b/img/benchmark_H100.png differ