diff --git a/tests/integration/test_default_tabular.py b/tests/integration/test_default_tabular.py
new file mode 100644
index 00000000..70d2286c
--- /dev/null
+++ b/tests/integration/test_default_tabular.py
@@ -0,0 +1,21 @@
+import pytest
+
+from sklearn.metrics import roc_auc_score
+
+from lightautoml.automl.presets.tabular_presets import TabularAutoML
+
+
+@pytest.mark.integtest
+def test_default_tabular(sampled_app_train_test, sampled_app_roles, binary_task):
+    # load and prepare data
+    train, test = sampled_app_train_test
+
+    # run automl
+    automl = TabularAutoML(task=binary_task)
+    oof_predictions = automl.fit_predict(train, roles=sampled_app_roles)
+    te_pred = automl.predict(test)
+
+    # calculate scores
+    print(f"Score for out-of-fold predictions: {roc_auc_score(train['TARGET'].values, oof_predictions.data[:, 0])}")
+    print(f"Score for hold-out: {roc_auc_score(test['TARGET'].values, te_pred.data[:, 0])}")
+    # add check