diff --git a/tests/brevitas/test_brevitas_relu_act_export.py b/tests/brevitas/test_brevitas_relu_act_export.py
index bb59a8414feffbb8362de629f8b30ac200a5227f..305382bbcaeef2d2c9e54523619255ddf1ff6d32 100644
--- a/tests/brevitas/test_brevitas_relu_act_export.py
+++ b/tests/brevitas/test_brevitas_relu_act_export.py
@@ -36,11 +36,14 @@ import torch
 from brevitas.core.quant import QuantType
 from brevitas.core.restrict_val import RestrictValueType
 from brevitas.core.scaling import ScalingImplType
+from brevitas.export.onnx.generic.manager import BrevitasONNXManager
 from brevitas.nn import QuantReLU
+from qonnx.util.cleanup import cleanup as qonnx_cleanup
 
 import finn.core.onnx_exec as oxe
 from finn.core.modelwrapper import ModelWrapper
 from finn.transformation.infer_shapes import InferShapes
+from finn.transformation.qonnx.convert_qonnx_to_finn import ConvertQONNXtoFINN
 
 export_onnx_path = "test_brevitas_relu_act_export.onnx"
 
@@ -50,7 +53,8 @@ export_onnx_path = "test_brevitas_relu_act_export.onnx"
 @pytest.mark.parametrize(
     "scaling_impl_type", [ScalingImplType.CONST, ScalingImplType.PARAMETER]
 )
-def test_brevitas_act_export_relu(abits, max_val, scaling_impl_type):
+@pytest.mark.parametrize("QONNX_export", [False, True])
+def test_brevitas_act_export_relu(abits, max_val, scaling_impl_type, QONNX_export):
     min_val = -1.0
     ishape = (1, 15)
 
@@ -71,8 +75,15 @@ scaling_impl.learned_value": torch.tensor(
             )
         }
         b_act.load_state_dict(checkpoint)
-
-    bo.export_finn_onnx(b_act, ishape, export_onnx_path)
+    if QONNX_export:
+        m_path = export_onnx_path
+        BrevitasONNXManager.export(b_act, ishape, m_path)
+        qonnx_cleanup(m_path, out_file=m_path)
+        model = ModelWrapper(m_path)
+        model = model.transform(ConvertQONNXtoFINN())
+        model.save(m_path)
+    else:
+        bo.export_finn_onnx(b_act, ishape, export_onnx_path)
     model = ModelWrapper(export_onnx_path)
     model = model.transform(InferShapes())
     inp_tensor = np.random.uniform(low=min_val, high=max_val, size=ishape).astype(