diff --git a/training/distributed_training/pytorch/model_parallel/flan-t5/requirements.txt b/training/distributed_training/pytorch/model_parallel/flan-t5/requirements.txt
index 67d35f169d..a7f48435ba 100644
--- a/training/distributed_training/pytorch/model_parallel/flan-t5/requirements.txt
+++ b/training/distributed_training/pytorch/model_parallel/flan-t5/requirements.txt
@@ -4,6 +4,6 @@ sagemaker
 sagemaker-experiments
 scipy
 torchnet
-transformers==4.21.0
+transformers==4.38.0
 smdebug
 humanize