diff --git a/tests/models/autoencoders/test_models_asymmetric_autoencoder_kl.py b/tests/models/autoencoders/test_models_asymmetric_autoencoder_kl.py
index 36207c6143..2476ab92f7 100644
--- a/tests/models/autoencoders/test_models_asymmetric_autoencoder_kl.py
+++ b/tests/models/autoencoders/test_models_asymmetric_autoencoder_kl.py
@@ -36,12 +36,13 @@ from ...testing_utils import (
     torch_device,
 )
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class AsymmetricAutoencoderKLTests(ModelTesterMixin, unittest.TestCase):
+class AsymmetricAutoencoderKLTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = AsymmetricAutoencoderKL
     main_input_name = "sample"
     base_precision = 1e-2
diff --git a/tests/models/autoencoders/test_models_autoencoder_cosmos.py b/tests/models/autoencoders/test_models_autoencoder_cosmos.py
index 45ae737860..5898ae776a 100644
--- a/tests/models/autoencoders/test_models_autoencoder_cosmos.py
+++ b/tests/models/autoencoders/test_models_autoencoder_cosmos.py
@@ -18,12 +18,13 @@ from diffusers import AutoencoderKLCosmos
 
 from ...testing_utils import enable_full_determinism, floats_tensor, torch_device
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class AutoencoderKLCosmosTests(ModelTesterMixin, unittest.TestCase):
+class AutoencoderKLCosmosTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = AutoencoderKLCosmos
     main_input_name = "sample"
     base_precision = 1e-2
@@ -80,7 +81,3 @@ class AutoencoderKLCosmosTests(ModelTesterMixin, unittest.TestCase):
     @unittest.skip("Not sure why this test fails. Investigate later.")
     def test_effective_gradient_checkpointing(self):
         pass
-
-    @unittest.skip("Unsupported test.")
-    def test_forward_with_norm_groups(self):
-        pass
diff --git a/tests/models/autoencoders/test_models_autoencoder_dc.py b/tests/models/autoencoders/test_models_autoencoder_dc.py
index 0b48818bd1..a6912f3eba 100644
--- a/tests/models/autoencoders/test_models_autoencoder_dc.py
+++ b/tests/models/autoencoders/test_models_autoencoder_dc.py
@@ -23,12 +23,13 @@ from ...testing_utils import (
     torch_device,
 )
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class AutoencoderDCTests(ModelTesterMixin, unittest.TestCase):
+class AutoencoderDCTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = AutoencoderDC
     main_input_name = "sample"
     base_precision = 1e-2
@@ -81,7 +82,3 @@ class AutoencoderDCTests(ModelTesterMixin, unittest.TestCase):
         init_dict = self.get_autoencoder_dc_config()
         inputs_dict = self.dummy_input
         return init_dict, inputs_dict
-
-    @unittest.skip("AutoencoderDC does not support `norm_num_groups` because it does not use GroupNorm.")
-    def test_forward_with_norm_groups(self):
-        pass
diff --git a/tests/models/autoencoders/test_models_autoencoder_kl_temporal_decoder.py b/tests/models/autoencoders/test_models_autoencoder_kl_temporal_decoder.py
index 66062e21e7..93f40f44a9 100644
--- a/tests/models/autoencoders/test_models_autoencoder_kl_temporal_decoder.py
+++ b/tests/models/autoencoders/test_models_autoencoder_kl_temporal_decoder.py
@@ -23,12 +23,13 @@ from ...testing_utils import (
     torch_device,
 )
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class AutoencoderKLTemporalDecoderTests(ModelTesterMixin, unittest.TestCase):
+class AutoencoderKLTemporalDecoderTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = AutoencoderKLTemporalDecoder
     main_input_name = "sample"
     base_precision = 1e-2
@@ -67,7 +68,3 @@ class AutoencoderKLTemporalDecoderTests(ModelTesterMixin, unittest.TestCase):
     def test_gradient_checkpointing_is_applied(self):
         expected_set = {"Encoder", "TemporalDecoder", "UNetMidBlock2D"}
         super().test_gradient_checkpointing_is_applied(expected_set=expected_set)
-
-    @unittest.skip("Test unsupported.")
-    def test_forward_with_norm_groups(self):
-        pass
diff --git a/tests/models/autoencoders/test_models_autoencoder_magvit.py b/tests/models/autoencoders/test_models_autoencoder_magvit.py
index f77564fbb0..fe35422e95 100644
--- a/tests/models/autoencoders/test_models_autoencoder_magvit.py
+++ b/tests/models/autoencoders/test_models_autoencoder_magvit.py
@@ -19,12 +19,13 @@ from diffusers import AutoencoderKLMagvit
 
 from ...testing_utils import enable_full_determinism, floats_tensor, torch_device
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class AutoencoderKLMagvitTests(ModelTesterMixin, unittest.TestCase):
+class AutoencoderKLMagvitTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = AutoencoderKLMagvit
     main_input_name = "sample"
     base_precision = 1e-2
diff --git a/tests/models/autoencoders/test_models_autoencoder_mochi.py b/tests/models/autoencoders/test_models_autoencoder_mochi.py
index 371f0dd68f..ab8d429a67 100755
--- a/tests/models/autoencoders/test_models_autoencoder_mochi.py
+++ b/tests/models/autoencoders/test_models_autoencoder_mochi.py
@@ -17,18 +17,15 @@ import unittest
 
 from diffusers import AutoencoderKLMochi
 
-from ...testing_utils import (
-    enable_full_determinism,
-    floats_tensor,
-    torch_device,
-)
+from ...testing_utils import enable_full_determinism, floats_tensor, torch_device
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class AutoencoderKLMochiTests(ModelTesterMixin, unittest.TestCase):
+class AutoencoderKLMochiTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = AutoencoderKLMochi
     main_input_name = "sample"
     base_precision = 1e-2
@@ -79,14 +76,6 @@ class AutoencoderKLMochiTests(ModelTesterMixin, unittest.TestCase):
         }
         super().test_gradient_checkpointing_is_applied(expected_set=expected_set)
 
-    @unittest.skip("Unsupported test.")
-    def test_forward_with_norm_groups(self):
-        """
-        tests/models/autoencoders/test_models_autoencoder_mochi.py::AutoencoderKLMochiTests::test_forward_with_norm_groups -
-        TypeError: AutoencoderKLMochi.__init__() got an unexpected keyword argument 'norm_num_groups'
-        """
-        pass
-
     @unittest.skip("Unsupported test.")
     def test_model_parallelism(self):
         """
diff --git a/tests/models/autoencoders/test_models_autoencoder_oobleck.py b/tests/models/autoencoders/test_models_autoencoder_oobleck.py
index debbb5fdeb..d10e8ba33a 100644
--- a/tests/models/autoencoders/test_models_autoencoder_oobleck.py
+++ b/tests/models/autoencoders/test_models_autoencoder_oobleck.py
@@ -31,12 +31,13 @@ from ...testing_utils import (
     torch_device,
 )
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class AutoencoderOobleckTests(ModelTesterMixin, unittest.TestCase):
+class AutoencoderOobleckTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = AutoencoderOobleck
     main_input_name = "sample"
     base_precision = 1e-2
@@ -106,10 +107,6 @@ class AutoencoderOobleckTests(ModelTesterMixin, unittest.TestCase):
             "Without slicing outputs should match with the outputs when slicing is manually disabled.",
         )
 
-    @unittest.skip("Test unsupported.")
-    def test_forward_with_norm_groups(self):
-        pass
-
     @unittest.skip("No attention module used in this model")
     def test_set_attn_processor_for_determinism(self):
         return
diff --git a/tests/models/autoencoders/test_models_vq.py b/tests/models/autoencoders/test_models_vq.py
index c92b022779..b88d24d1f2 100644
--- a/tests/models/autoencoders/test_models_vq.py
+++ b/tests/models/autoencoders/test_models_vq.py
@@ -19,19 +19,15 @@ import torch
 
 from diffusers import VQModel
 
-from ...testing_utils import (
-    backend_manual_seed,
-    enable_full_determinism,
-    floats_tensor,
-    torch_device,
-)
+from ...testing_utils import backend_manual_seed, enable_full_determinism, floats_tensor, torch_device
 from ..test_modeling_common import ModelTesterMixin
+from .testing_utils import AutoencoderTesterMixin
 
 
 enable_full_determinism()
 
 
-class VQModelTests(ModelTesterMixin, unittest.TestCase):
+class VQModelTests(ModelTesterMixin, AutoencoderTesterMixin, unittest.TestCase):
     model_class = VQModel
     main_input_name = "sample"