diff --git a/forge/test/mlir/llama/test_llama_inference.py b/forge/test/mlir/llama/test_llama_inference.py
index 38ad8a9b7..af439c2f6 100644
--- a/forge/test/mlir/llama/test_llama_inference.py
+++ b/forge/test/mlir/llama/test_llama_inference.py
@@ -11,8 +11,8 @@
 
 
 @pytest.mark.parametrize("model_path", ["openlm-research/open_llama_3b", "meta-llama/Llama-3.2-1B"])
-@pytest.mark.xfail()
 @pytest.mark.push
+@pytest.mark.skip(reason="Out of system memory during compile time. Skipping until resolved")
 def test_llama_inference(model_path):
     if model_path == "meta-llama/Llama-3.2-1B":
         pytest.skip("Skipping test for Llama-3.2-1B model, waiting for new transformers version.")