diff --git a/README.md b/README.md
index 74182bc09320408ee80dbbfda116b2f334660199..01b178bff9c26a70c865677f1dca1beec00d141f 100644
--- a/README.md
+++ b/README.md
@@ -1005,5 +1005,9 @@ Download [NCCL2](https://developer.nvidia.com/nccl/nccl-download), and export th
 ### FAQ6: Runtime error: `ModuleNotFoundError No module named propeller`<a name="faq6"></a>
 
 you can import propeller to your PYTHONPATH by `export PYTHONPATH:./:$PYTHONPATH`
-`
+
+
+#### FAQ7: Cannot malloc XXX MB GPU memory.
+
+Try to reduce the batch_size, reduce the max_seq_len and set FLAGS_eager_delete_tensor_gb=0.0
 
diff --git a/README.zh.md b/README.zh.md
index 9fce0f3ddf6ceceb6ff12ea486fa90a0b8e339b5..ed5ea5fb3f1759bd5e964d3b8dbb40b52458937a 100644
--- a/README.zh.md
+++ b/README.zh.md
@@ -1136,3 +1136,7 @@ python -u infer_classifyer.py \
 ### FAQ6: 运行报错`ModuleNotFoundError: No module named 'propeller'`<a name="faq6"></a>
 
 您可以通过`export PYTHONPATH=./:$PYTHONPATH`的方式引入Propeller.
+
+### FAQ7：显存不足报错: Cannot malloc XXX MB GPU memory. 
+
+您可以通过减小batch_size, 减小max_seq_len, 设置FLAGS_eager_delete_tensor_gb=0.0来减小网络对显存的需求