Fix CI rate limiting (#1449)

* fix CI

* fix

Author: IlyasMoutawwakil

.github/workflows/test_inc.yml

@@ -15,6 +15,7 @@ concurrency:
 
 env:
   TRANSFORMERS_IS_CI: true
+  HF_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
 
 jobs:
   build:
@@ -48,5 +49,4 @@ jobs:
       - name: Test with Pytest
         run: |
           pytest tests/neural_compressor
-        env:
-          HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
+

.github/workflows/test_ipex.yml

@@ -15,6 +15,7 @@ concurrency:
 
 env:
   TRANSFORMERS_IS_CI: true
+  HF_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
 
 jobs:
   build:

.github/workflows/test_openvino.yml

@@ -21,7 +21,7 @@ env:
   UV_TORCH_BACKEND: cpu
   UV_SYSTEM_PYTHON: true
   TRANSFORMERS_IS_CI: true
-  HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
+  HF_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
 
 jobs:
   build:

.github/workflows/test_openvino_nightly.yml

@@ -27,7 +27,7 @@ env:
   UV_TORCH_BACKEND: cpu
   UV_SYSTEM_PYTHON: true
   TRANSFORMERS_IS_CI: true
-  HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
+  HF_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
 
 jobs:
   build:

.github/workflows/test_openvino_notebooks.yml

@@ -18,6 +18,7 @@ concurrency:
 
 env:
   TRANSFORMERS_IS_CI: true
+  HF_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
 
 jobs:
   build:

.github/workflows/test_openvino_slow.yml

@@ -24,7 +24,7 @@ env:
   UV_TORCH_BACKEND: cpu
   UV_SYSTEM_PYTHON: true
   TRANSFORMERS_IS_CI: true
-  HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
+  HF_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
 
 jobs:
   build:

optimum/intel/openvino/modeling_sentence_transformers.py

@@ -153,3 +153,6 @@ def tokenize(
             )
         )
         return output
+
+    def get_model_kwargs(self):
+        return []

tests/openvino/test_modeling.py

@@ -572,9 +572,9 @@ def test_load_from_hub_and_save_text_speech_model(self):
     @slow
     def test_load_model_from_hub_private_with_token(self):
         model_id = "optimum-internal-testing/tiny-random-phi-private"
-        token = os.environ.get("HF_HUB_READ_TOKEN", None)
+        token = os.environ.get("HF_TOKEN", None)
         if not token:
-            self.skipTest("Test requires a token `HF_HUB_READ_TOKEN` in the environment variable")
+            self.skipTest("Test requires a token `HF_TOKEN` in the environment variable")
 
         model = OVModelForCausalLM.from_pretrained(model_id, token=token, revision="openvino")
         self.assertIsInstance(model.config, PretrainedConfig)