diff --git a/tests/generation/test_utils.py b/tests/generation/test_utils.py
index 384c47fef47..3204c011a63 100644
--- a/tests/generation/test_utils.py
+++ b/tests/generation/test_utils.py
@@ -34,6 +34,7 @@ from transformers.testing_utils import (
     require_accelerate,
     require_flash_attn,
     require_optimum_quanto,
+    require_read_token,
     require_torch,
     require_torch_accelerator,
     require_torch_gpu,
@@ -4283,6 +4284,8 @@ class GenerationIntegrationTests(unittest.TestCase):
         gen_out = compiled_generate(**model_inputs, generation_config=generation_config)
         self.assertTrue(gen_out.shape[1] > model_inputs["input_ids"].shape[1])  # some text was generated
 
+    @require_read_token
+    @slow
     def test_assisted_generation_early_exit(self):
         """
         Tests that assisted generation with early exit works as expected. Under the hood, this has complex cache
@@ -4791,6 +4794,7 @@ class GenerationIntegrationTests(unittest.TestCase):
         self.assertTrue(np.array_equal(output_sequences_decoder_input_ids, output_sequences_input_ids))
         self.assertTrue(np.array_equal(output_sequences_decoder_input_ids[:, 1:2], conditioning_input))
 
+    @require_read_token
     @slow
     @require_torch_gpu
     def test_cache_device_map_with_vision_layer_device_map(self):