diff --git a/docs/examples/config.rst b/docs/examples/config.rst
index ec6006ec3bf..2f27e448792 100644
--- a/docs/examples/config.rst
+++ b/docs/examples/config.rst
@@ -509,6 +509,13 @@ Trainer
   for the ray register center to be ready. Default is 300 seconds.
 
 
+This figure illustrates how the configurations affect the training.
+
+https://excalidraw.com/#json=pfhkRmiLm1jnnRli9VFhb,Ut4E8peALlgAUpr7E5pPCA
+
+.. image:: https://github.com/user-attachments/assets/16aebad1-0da6-4eb3-806d-54a74e712c2d
+
+
 evaluation.yaml
 ---------------
 
diff --git a/docs/faq/faq.rst b/docs/faq/faq.rst
index 5cd555fd481..c836b0613fc 100644
--- a/docs/faq/faq.rst
+++ b/docs/faq/faq.rst
@@ -107,6 +107,8 @@ https://verl.readthedocs.io/en/latest/examples/config.html to disable just-in-ti
 What is the meaning of train batch size, mini batch size, and micro batch size?
 ------------------------------------------------------------------------------------------
 
-Please check out the following figure from the community (credit to @hiyouga)
+This figure illustrates the relationship between different batch size configurations.
 
-.. image:: https://github.com/hiyouga/EasyR1/blob/main/assets/easyr1_grpo.png
+https://excalidraw.com/#json=pfhkRmiLm1jnnRli9VFhb,Ut4E8peALlgAUpr7E5pPCA
+
+.. image:: https://github.com/user-attachments/assets/16aebad1-0da6-4eb3-806d-54a74e712c2d