Some minor updates to the nightly and push workflows (#9759)

* move lora integration tests to nightly./

* remove slow marker in the workflow where not needed.

Author: sayakpaul

.github/workflows/push_tests.yml

@@ -81,7 +81,7 @@ jobs:
       - name: Environment
         run: |
           python utils/print_env.py
-      - name: Slow PyTorch CUDA checkpoint tests on Ubuntu
+      - name: PyTorch CUDA checkpoint tests on Ubuntu
         env:
           HF_TOKEN: ${{ secrets.HF_TOKEN }}
           # https://pytorch.org/docs/stable/notes/randomness.html#avoiding-nondeterministic-algorithms
@@ -184,7 +184,7 @@ jobs:
       run: |
         python utils/print_env.py
 
-    - name: Run slow Flax TPU tests
+    - name: Run Flax TPU tests
       env:
         HF_TOKEN: ${{ secrets.HF_TOKEN }}
       run: |
@@ -232,7 +232,7 @@ jobs:
       run: |
         python utils/print_env.py
 
-    - name: Run slow ONNXRuntime CUDA tests
+    - name: Run ONNXRuntime CUDA tests
       env:
         HF_TOKEN: ${{ secrets.HF_TOKEN }}
       run: |

tests/lora/test_lora_layers_flux.py

@@ -27,6 +27,7 @@
 from diffusers.utils.testing_utils import (
     floats_tensor,
     is_peft_available,
+    nightly,
     numpy_cosine_similarity_distance,
     require_peft_backend,
     require_torch_gpu,
@@ -165,9 +166,10 @@ def test_modify_padding_mode(self):
 
 
 @slow
+@nightly
 @require_torch_gpu
 @require_peft_backend
-# @unittest.skip("We cannot run inference on this model with the current CI hardware")
+@unittest.skip("We cannot run inference on this model with the current CI hardware")
 # TODO (DN6, sayakpaul): move these tests to a beefier GPU
 class FluxLoRAIntegrationTests(unittest.TestCase):
     """internal note: The integration slices were obtained on audace.

tests/lora/test_lora_layers_sd.py

@@ -34,6 +34,7 @@
 from diffusers.utils.import_utils import is_accelerate_available
 from diffusers.utils.testing_utils import (
     load_image,
+    nightly,
     numpy_cosine_similarity_distance,
     require_peft_backend,
     require_torch_gpu,
@@ -207,6 +208,7 @@ def test_integration_move_lora_dora_cpu(self):
 
 
 @slow
+@nightly
 @require_torch_gpu
 @require_peft_backend
 class LoraIntegrationTests(unittest.TestCase):

tests/lora/test_lora_layers_sdxl.py

@@ -113,6 +113,7 @@ def tearDown(self):
 
 
 @slow
+@nightly
 @require_torch_gpu
 @require_peft_backend
 class LoraSDXLIntegrationTests(unittest.TestCase):