diff --git a/tests/models/mllama/test_modeling_mllama.py b/tests/models/mllama/test_modeling_mllama.py
index e67e0455e1f..589cff4c02b 100644
--- a/tests/models/mllama/test_modeling_mllama.py
+++ b/tests/models/mllama/test_modeling_mllama.py
@@ -17,7 +17,6 @@ import unittest
 
 import pytest
 import requests
-from parameterized import parameterized
 
 from transformers import (
     AutoProcessor,
@@ -396,12 +395,6 @@ class MllamaForConditionalGenerationModelTest(ModelTesterMixin, GenerationTester
     def test_model_parallelism(self):
         pass
 
-    @parameterized.expand([("offloaded",)])
-    @pytest.mark.generate
-    @unittest.skip(reason="Offloaded cache seems to not work with mllama's kv cache type")
-    def test_offloaded_cache_implementation(self, cache_implementation):
-        pass
-
     @unittest.skip(
         reason="Mllama cache type doesn't allow correct check on output `past_key_values` due to `Cache.crop()`"
     )
diff --git a/tests/models/whisper/test_modeling_whisper.py b/tests/models/whisper/test_modeling_whisper.py
index 161bb33a801..ffc4b59abb7 100644
--- a/tests/models/whisper/test_modeling_whisper.py
+++ b/tests/models/whisper/test_modeling_whisper.py
@@ -546,12 +546,6 @@ class WhisperModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMi
     def test_generate_with_head_masking(self):
         pass
 
-    @parameterized.expand([("offloaded",)])
-    @pytest.mark.generate
-    @unittest.skip(reason="Whisper doesn't work with offloaded cache implementation yet")
-    def test_offloaded_cache_implementation(self, cache_implementation):
-        pass
-
     @require_torch_fp16
     def test_generate_fp16(self):
         config, input_dict = self.model_tester.prepare_config_and_inputs()