diff --git a/examples/llm_eval/lm_eval_hf.py b/examples/llm_eval/lm_eval_hf.py
index ce3b8e1442..dbdf22d868 100755
--- a/examples/llm_eval/lm_eval_hf.py
+++ b/examples/llm_eval/lm_eval_hf.py
@@ -40,8 +40,15 @@
 import warnings
 
 import datasets
+import lm_eval
 from lm_eval import utils
 from lm_eval.__main__ import cli_evaluate, parse_eval_args, setup_parser
+
+if not lm_eval.__version__.startswith("0.4.8"):
+    warnings.warn(
+        f"lm_eval_hf.py is tested with lm-eval 0.4.8; found {lm_eval.__version__}. "
+        "Later versions may have incompatible API changes."
+    )
 from lm_eval.api.model import T
 from lm_eval.models.huggingface import HFLM
 from quantization_utils import quantize_model
diff --git a/examples/puzzletron/requirements.txt b/examples/puzzletron/requirements.txt
index fd227a2544..317a38f5ea 100644
--- a/examples/puzzletron/requirements.txt
+++ b/examples/puzzletron/requirements.txt
@@ -1,4 +1,4 @@
-lm-eval==0.4.10
+lm-eval==0.4.8
 math-verify
 ray
 # Likely works for transformers v5 also, but we need to test it