diff --git a/ci/test_wheel.sh b/ci/test_wheel.sh
index 47ad8e0b1b..a0adddc6b7 100755
--- a/ci/test_wheel.sh
+++ b/ci/test_wheel.sh
@@ -15,30 +15,29 @@ fi
 # echo to expand wildcard before adding `[extra]` requires for pip
 python -m pip install $(echo ./dist/cuml*.whl)[test]
 
+RAPIDS_TESTS_DIR=${RAPIDS_TESTS_DIR:-"${PWD}/test-results"}
+mkdir -p "${RAPIDS_TESTS_DIR}"
+
 EXITCODE=0
 trap "EXITCODE=1" ERR
 set +e
 
-# Run smoke tests for aarch64 pull requests
-if [[ "$(arch)" == "aarch64" && "${RAPIDS_BUILD_TYPE}" == "pull-request" ]]; then
-    python ci/wheel_smoke_test_cuml.py
-else
-    rapids-logger "pytest cuml single GPU"
-    ./ci/run_cuml_singlegpu_pytests.sh \
-      --numprocesses=8 \
-      --dist=worksteal \
-      -k 'not test_sparse_pca_inputs and not test_fil_skl_classification' \
-      --junitxml="${RAPIDS_TESTS_DIR}/junit-cuml.xml"
-
-    # Run test_sparse_pca_inputs separately
-    ./ci/run_cuml_singlegpu_pytests.sh \
-      -k 'test_sparse_pca_inputs' \
-      --junitxml="${RAPIDS_TESTS_DIR}/junit-cuml-sparse-pca.xml"
-
-    rapids-logger "pytest cuml-dask"
-    ./ci/run_cuml_dask_pytests.sh \
-      --junitxml="${RAPIDS_TESTS_DIR}/junit-cuml-dask.xml"
-fi
+
+rapids-logger "pytest cuml single GPU"
+./ci/run_cuml_singlegpu_pytests.sh \
+  --numprocesses=8 \
+  --dist=worksteal \
+  -k 'not test_sparse_pca_inputs and not test_fil_skl_classification' \
+  --junitxml="${RAPIDS_TESTS_DIR}/junit-cuml.xml"
+
+# Run test_sparse_pca_inputs separately
+./ci/run_cuml_singlegpu_pytests.sh \
+  -k 'test_sparse_pca_inputs' \
+  --junitxml="${RAPIDS_TESTS_DIR}/junit-cuml-sparse-pca.xml"
+
+rapids-logger "pytest cuml-dask"
+./ci/run_cuml_dask_pytests.sh \
+  --junitxml="${RAPIDS_TESTS_DIR}/junit-cuml-dask.xml"
 
 rapids-logger "Test script exiting with value: $EXITCODE"
 exit ${EXITCODE}
diff --git a/ci/wheel_smoke_test_cuml.py b/ci/wheel_smoke_test_cuml.py
deleted file mode 100644
index aa935639ae..0000000000
--- a/ci/wheel_smoke_test_cuml.py
+++ /dev/null
@@ -1,26 +0,0 @@
-"""
-A simple test for cuML based on scikit-learn.
-
-Adapted from xgboost:
-https://raw.githubusercontent.com/rapidsai/xgboost-conda/branch-23.02/recipes/xgboost/test-py-xgboost.py
-"""
-from cuml.ensemble import RandomForestClassifier
-import sklearn.datasets
-import sklearn.model_selection
-import sklearn.metrics
-
-X, y = sklearn.datasets.load_iris(return_X_y=True)
-Xtrn, Xtst, ytrn, ytst = sklearn.model_selection.train_test_split(
-    X, y, train_size=0.8, random_state=4)
-
-clf = RandomForestClassifier(
-    max_depth=2,
-    n_estimators=10,
-    n_bins=32,
-    accuracy_metric="multi:softmax")
-clf.fit(Xtrn, ytrn)
-ypred = clf.predict(Xtst)
-acc = sklearn.metrics.accuracy_score(ytst, ypred)
-
-print('cuml RandomForestClassifier accuracy on iris:', acc)
-assert acc > 0.9