diff --git a/tests/entrypoints/llm/test_accuracy.py b/tests/entrypoints/llm/test_accuracy.py
index a2d35486a5e8..7e6bd3664ebd 100644
--- a/tests/entrypoints/llm/test_accuracy.py
+++ b/tests/entrypoints/llm/test_accuracy.py
@@ -69,6 +69,12 @@ def test_lm_eval_accuracy_v1_engine(model, monkeypatch: pytest.MonkeyPatch):
         more_args = None
         if current_platform.is_tpu():
             # Limit compilation time for TPU V1
+
+            if model == "google/gemma-3-1b-it":
+                pytest.skip(
+                    "Temporarily disabled due to test failures"
+                    "(timeout or accuracy mismatch). Re-enable once fixed.")
+
             more_args = "max_model_len=2048,max_num_seqs=64"
 
             # Add TP test (if provided)