diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
index c15a3473..32a7492d 100644
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@@ -19,8 +19,8 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [windows-latest] # ubuntu-latest, macos-latest
-        python-version: ["3.11"] # -> Will re-enable support for py312 once pyg is released, "3.10",
+        os: [ubuntu-latest, macos-latest, windows-latest]
+        python-version: ["3.10", "3.11"] # -> Will re-enable support for py312 once pyg is released, "3.10",
 
     runs-on: ${{ matrix.os }}
     timeout-minutes: 30
diff --git a/tests/test_datamodule.py b/tests/test_datamodule.py
index 18e90b79..31e8c4f8 100644
--- a/tests/test_datamodule.py
+++ b/tests/test_datamodule.py
@@ -11,13 +11,13 @@
 --------------------------------------------------------------------------------
 """
 
+import os
 import unittest as ut
 import numpy as np
 import torch
 import pandas as pd
 import tempfile
 
-import graphium
 from graphium.utils.fs import rm, exists, get_size
 from graphium.data import GraphOGBDataModule, MultitaskFromSmilesDataModule
 
@@ -25,8 +25,8 @@
 
 TEMP_CACHE_DATA_PATH = "tests/temp_cache_0000"
 
-
 class test_DataModule(ut.TestCase):
+    
     def test_ogb_datamodule(self):
         # other datasets are too large to be tested
         dataset_names = ["ogbg-molhiv", "ogbg-molpcba", "ogbg-moltox21", "ogbg-molfreesolv"]
@@ -380,7 +380,7 @@ def test_datamodule_multiple_data_files(self):
 
         self.assertEqual(len(ds.train_ds), 20)
 
-    def test_splits_file(self, tmp_path):
+    def test_splits_file(self):
         # Test single CSV files
         csv_file = "tests/data/micro_ZINC_shard_1.csv"
         df = pd.read_csv(csv_file)
@@ -423,15 +423,17 @@ def test_splits_file(self, tmp_path):
         self.assertEqual(len(ds.val_ds), len(split_val))
         self.assertEqual(len(ds.test_ds), len(split_test))
 
-        # Create a TemporaryFile to save the splits, and test the datamodule
-        with tempfile.NamedTemporaryFile(suffix=".pt", dir=tmp_path) as temp:
+        try: 
+            # Create a TemporaryFile to save the splits, and test the datamodule
+            temp_file = tempfile.NamedTemporaryFile(suffix=".pt", delete=False)
+            
             # Save the splits
-            torch.save(splits, temp)
+            torch.save(splits, temp_file)
 
             # Test the datamodule
             task_kwargs = {
                 "df_path": csv_file,
-                "splits_path": temp.name,
+                "splits_path": temp_file.name,
                 "split_val": 0.0,
                 "split_test": 0.0,
             }
@@ -468,6 +470,10 @@ def test_splits_file(self, tmp_path):
             )
             np.testing.assert_array_equal(ds.val_ds.smiles_offsets_tensor, ds2.val_ds.smiles_offsets_tensor)
             np.testing.assert_array_equal(ds.test_ds.smiles_offsets_tensor, ds2.test_ds.smiles_offsets_tensor)
+        
+        finally:
+            temp_file.close()
+            os.unlink(temp_file.name)
 
 
 if __name__ == "__main__":