diff --git a/tests/benchmark/read_gbq_colab/aggregate_output.py b/tests/benchmark/read_gbq_colab/aggregate_output.py
index dda4bf95a4..0086e97ee2 100644
--- a/tests/benchmark/read_gbq_colab/aggregate_output.py
+++ b/tests/benchmark/read_gbq_colab/aggregate_output.py
@@ -15,17 +15,15 @@
 
 import benchmark.utils as utils
 
-import bigframes.session
+import bigframes.pandas
 
 PAGE_SIZE = utils.READ_GBQ_COLAB_PAGE_SIZE
 
 
-def aggregate_output(
-    *, project_id, dataset_id, table_id, session: bigframes.session.Session
-):
+def aggregate_output(*, project_id, dataset_id, table_id):
     # TODO(tswast): Support alternative query if table_id is a local DataFrame,
     # e.g. "{local_inline}" or "{local_large}"
-    df = session._read_gbq_colab(
+    df = bigframes.pandas._read_gbq_colab(
         f"SELECT * FROM `{project_id}`.{dataset_id}.{table_id}"
     )
 
@@ -56,9 +54,8 @@ def aggregate_output(
         project_id,
         dataset_id,
         table_id,
-        session,
         suffix,
-    ) = utils.get_configuration(include_table_id=True)
+    ) = utils.get_configuration(include_table_id=True, include_session=False)
     current_path = pathlib.Path(__file__).absolute()
 
     utils.get_execution_time(
@@ -68,5 +65,4 @@ def aggregate_output(
         project_id=project_id,
         dataset_id=dataset_id,
         table_id=table_id,
-        session=session,
     )
diff --git a/tests/benchmark/read_gbq_colab/dry_run.py b/tests/benchmark/read_gbq_colab/dry_run.py
index c2de1b7cc4..dedbf2c9e2 100644
--- a/tests/benchmark/read_gbq_colab/dry_run.py
+++ b/tests/benchmark/read_gbq_colab/dry_run.py
@@ -15,13 +15,13 @@
 
 import benchmark.utils as utils
 
-import bigframes.session
+import bigframes.pandas
 
 
-def dry_run(*, project_id, dataset_id, table_id, session: bigframes.session.Session):
+def dry_run(*, project_id, dataset_id, table_id):
     # TODO(tswast): Support alternative query if table_id is a local DataFrame,
     # e.g. "{local_inline}" or "{local_large}"
-    session._read_gbq_colab(
+    bigframes.pandas._read_gbq_colab(
         f"SELECT * FROM `{project_id}`.{dataset_id}.{table_id}",
         dry_run=True,
     )
@@ -32,9 +32,8 @@ def dry_run(*, project_id, dataset_id, table_id, session: bigframes.session.Sess
         project_id,
         dataset_id,
         table_id,
-        session,
         suffix,
-    ) = utils.get_configuration(include_table_id=True)
+    ) = utils.get_configuration(include_table_id=True, include_session=False)
     current_path = pathlib.Path(__file__).absolute()
 
     utils.get_execution_time(
@@ -44,5 +43,4 @@ def dry_run(*, project_id, dataset_id, table_id, session: bigframes.session.Sess
         project_id=project_id,
         dataset_id=dataset_id,
         table_id=table_id,
-        session=session,
     )
diff --git a/tests/benchmark/read_gbq_colab/filter_output.py b/tests/benchmark/read_gbq_colab/filter_output.py
index 0db7ac5fd6..f39dc28f7a 100644
--- a/tests/benchmark/read_gbq_colab/filter_output.py
+++ b/tests/benchmark/read_gbq_colab/filter_output.py
@@ -15,17 +15,15 @@
 
 import benchmark.utils as utils
 
-import bigframes.session
+import bigframes.pandas
 
 PAGE_SIZE = utils.READ_GBQ_COLAB_PAGE_SIZE
 
 
-def filter_output(
-    *, project_id, dataset_id, table_id, session: bigframes.session.Session
-):
+def filter_output(*, project_id, dataset_id, table_id):
     # TODO(tswast): Support alternative query if table_id is a local DataFrame,
     # e.g. "{local_inline}" or "{local_large}"
-    df = session._read_gbq_colab(
+    df = bigframes.pandas._read_gbq_colab(
         f"SELECT * FROM `{project_id}`.{dataset_id}.{table_id}"
     )
 
@@ -48,9 +46,8 @@ def filter_output(
         project_id,
         dataset_id,
         table_id,
-        session,
         suffix,
-    ) = utils.get_configuration(include_table_id=True)
+    ) = utils.get_configuration(include_table_id=True, include_session=False)
     current_path = pathlib.Path(__file__).absolute()
 
     utils.get_execution_time(
@@ -60,5 +57,4 @@ def filter_output(
         project_id=project_id,
         dataset_id=dataset_id,
         table_id=table_id,
-        session=session,
     )
diff --git a/tests/benchmark/read_gbq_colab/first_page.py b/tests/benchmark/read_gbq_colab/first_page.py
index 2df9990d22..8da8a91652 100644
--- a/tests/benchmark/read_gbq_colab/first_page.py
+++ b/tests/benchmark/read_gbq_colab/first_page.py
@@ -15,15 +15,15 @@
 
 import benchmark.utils as utils
 
-import bigframes.session
+import bigframes.pandas
 
 PAGE_SIZE = utils.READ_GBQ_COLAB_PAGE_SIZE
 
 
-def first_page(*, project_id, dataset_id, table_id, session: bigframes.session.Session):
+def first_page(*, project_id, dataset_id, table_id):
     # TODO(tswast): Support alternative query if table_id is a local DataFrame,
     # e.g. "{local_inline}" or "{local_large}"
-    df = session._read_gbq_colab(
+    df = bigframes.pandas._read_gbq_colab(
         f"SELECT * FROM `{project_id}`.{dataset_id}.{table_id}"
     )
 
@@ -37,9 +37,8 @@ def first_page(*, project_id, dataset_id, table_id, session: bigframes.session.S
         project_id,
         dataset_id,
         table_id,
-        session,
         suffix,
-    ) = utils.get_configuration(include_table_id=True)
+    ) = utils.get_configuration(include_table_id=True, include_session=False)
     current_path = pathlib.Path(__file__).absolute()
 
     utils.get_execution_time(
@@ -49,5 +48,4 @@ def first_page(*, project_id, dataset_id, table_id, session: bigframes.session.S
         project_id=project_id,
         dataset_id=dataset_id,
         table_id=table_id,
-        session=session,
     )
diff --git a/tests/benchmark/read_gbq_colab/last_page.py b/tests/benchmark/read_gbq_colab/last_page.py
index ad785a29e8..6f58908c67 100644
--- a/tests/benchmark/read_gbq_colab/last_page.py
+++ b/tests/benchmark/read_gbq_colab/last_page.py
@@ -15,15 +15,15 @@
 
 import benchmark.utils as utils
 
-import bigframes.session
+import bigframes.pandas
 
 PAGE_SIZE = utils.READ_GBQ_COLAB_PAGE_SIZE
 
 
-def last_page(*, project_id, dataset_id, table_id, session: bigframes.session.Session):
+def last_page(*, project_id, dataset_id, table_id):
     # TODO(tswast): Support alternative query if table_id is a local DataFrame,
     # e.g. "{local_inline}" or "{local_large}"
-    df = session._read_gbq_colab(
+    df = bigframes.pandas._read_gbq_colab(
         f"SELECT * FROM `{project_id}`.{dataset_id}.{table_id}"
     )
 
@@ -38,9 +38,8 @@ def last_page(*, project_id, dataset_id, table_id, session: bigframes.session.Se
         project_id,
         dataset_id,
         table_id,
-        session,
         suffix,
-    ) = utils.get_configuration(include_table_id=True)
+    ) = utils.get_configuration(include_table_id=True, include_session=False)
     current_path = pathlib.Path(__file__).absolute()
 
     utils.get_execution_time(
@@ -50,5 +49,4 @@ def last_page(*, project_id, dataset_id, table_id, session: bigframes.session.Se
         project_id=project_id,
         dataset_id=dataset_id,
         table_id=table_id,
-        session=session,
     )
diff --git a/tests/benchmark/read_gbq_colab/sort_output.py b/tests/benchmark/read_gbq_colab/sort_output.py
index 997de5683d..bf826eb645 100644
--- a/tests/benchmark/read_gbq_colab/sort_output.py
+++ b/tests/benchmark/read_gbq_colab/sort_output.py
@@ -15,17 +15,15 @@
 
 import benchmark.utils as utils
 
-import bigframes.session
+import bigframes.pandas
 
 PAGE_SIZE = utils.READ_GBQ_COLAB_PAGE_SIZE
 
 
-def sort_output(
-    *, project_id, dataset_id, table_id, session: bigframes.session.Session
-):
+def sort_output(*, project_id, dataset_id, table_id):
     # TODO(tswast): Support alternative query if table_id is a local DataFrame,
     # e.g. "{local_inline}" or "{local_large}"
-    df = session._read_gbq_colab(
+    df = bigframes.pandas._read_gbq_colab(
         f"SELECT * FROM `{project_id}`.{dataset_id}.{table_id}"
     )
 
@@ -48,9 +46,8 @@ def sort_output(
         project_id,
         dataset_id,
         table_id,
-        session,
         suffix,
-    ) = utils.get_configuration(include_table_id=True)
+    ) = utils.get_configuration(include_table_id=True, include_session=False)
     current_path = pathlib.Path(__file__).absolute()
 
     utils.get_execution_time(
@@ -60,5 +57,4 @@ def sort_output(
         project_id=project_id,
         dataset_id=dataset_id,
         table_id=table_id,
-        session=session,
     )