Cache: models return input cache type (#30716)

tests/models/olmo/test_modeling_olmo.py

@@ -353,11 +353,6 @@ class OlmoModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMixin
         # The output should be different for long inputs
         self.assertFalse(torch.allclose(original_long_output, scaled_long_output, atol=1e-5))
 
-    @unittest.skip("TODO @gante fix this for OLMo")
-    @parameterized.expand([(1, False), (1, True), (4, False)])
-    def test_new_cache_format(self, num_beams, do_sample):
-        pass
-
 
 @require_torch
 class OlmoIntegrationTest(unittest.TestCase):