diff --git a/docs/assets/IRgraph_markstep.png b/docs/assets/IRgraph_markstep.png
index a99895b47f5..2a9ad5ce54f 100644
Binary files a/docs/assets/IRgraph_markstep.png and b/docs/assets/IRgraph_markstep.png differ
diff --git a/docs/assets/IRgraph_no_markstep.png b/docs/assets/IRgraph_no_markstep.png
index 74f7b41cee4..282d3907104 100644
Binary files a/docs/assets/IRgraph_no_markstep.png and b/docs/assets/IRgraph_no_markstep.png differ
diff --git a/docs/assets/pytorchXLA_flow.svg b/docs/assets/pytorchXLA_flow.svg
index 0dd3bc96028..3812141ce48 100644
--- a/docs/assets/pytorchXLA_flow.svg
+++ b/docs/assets/pytorchXLA_flow.svg
@@ -1 +1 @@
-
\ No newline at end of file
+
diff --git a/docs/assets/spmd_debug_1_light.png b/docs/assets/spmd_debug_1_light.png
new file mode 100644
index 00000000000..9f2f060b2d0
Binary files /dev/null and b/docs/assets/spmd_debug_1_light.png differ
diff --git a/docs/assets/spmd_debug_2_light.png b/docs/assets/spmd_debug_2_light.png
new file mode 100644
index 00000000000..87deb04ce43
Binary files /dev/null and b/docs/assets/spmd_debug_2_light.png differ
diff --git a/docs/spmd.md b/docs/spmd.md
index 00e384e496c..ccd79ef1d35 100644
--- a/docs/spmd.md
+++ b/docs/spmd.md
@@ -501,7 +501,10 @@ xs.mark_sharding(t, mesh, ('x', 'y'))
from torch_xla.distributed.spmd.debugging import visualize_tensor_sharding
generated_table = visualize_tensor_sharding(t, use_color=False)
```
-
+
+
+
+
- Code snippet used `visualize_sharding` and visualization result:
@@ -510,7 +513,10 @@ from torch_xla.distributed.spmd.debugging import visualize_sharding
sharding = '{devices=[2,2]0,1,2,3}'
generated_table = visualize_sharding(sharding, use_color=False)
```
-
+
+
+
+
You could use these examples on TPU/GPU/CPU single-host and modify it to run on multi-host. And you could modify it to sharding-style `tiled`, `partial_replication` and `replicated`.
+
+