diff --git a/tests/models/test_llama.py b/tests/models/test_llama.py
new file mode 100644
index 00000000..1c52c15a
--- /dev/null
+++ b/tests/models/test_llama.py
@@ -0,0 +1,15 @@
+from unittest import TestCase
+
+from parameterized import parameterized
+from optimum.nvidia.utils.tests import requires_gpu
+from optimum.nvidia.models.llama import LLamaForCausalLM as TrtLlamaForCausalLM
+
+
+
+class LLamaForCausalLMTestCase(TestCase):
+
+    @requires_gpu
+    @parameterized.expand(["float16", "bfloat16"])
+    def test_build_engine_7b_with_tp(self, dtype: str):
+        model = TrtLlamaForCausalLM.from_pretrained("huggingface/llama-7b", dtype=dtype)
+        self.assertIsNotNone(model)
\ No newline at end of file