diff --git a/jraph/_src/models.py b/jraph/_src/models.py
index e6a2562..7135642 100644
--- a/jraph/_src/models.py
+++ b/jraph/_src/models.py
@@ -22,6 +22,7 @@
 import jax.tree_util as tree
 from jraph._src import graph as gn_graph
 from jraph._src import utils
+from contextlib import nullcontext
 
 # As of 04/2020 pytype doesn't support recursive types.
 # pytype: disable=not-supported-yet
@@ -548,18 +549,18 @@ def _ApplyGCN(graph):
     # Equivalent to jnp.sum(n_node), but jittable
     total_num_nodes = tree.tree_leaves(nodes)[0].shape[0]
     if add_self_edges:
-      # We add self edges to the senders and receivers so that each node
-      # includes itself in aggregation.
-      # In principle, a `GraphsTuple` should partition by n_edge, but in
-      # this case it is not required since a GCN is agnostic to whether
-      # the `GraphsTuple` is a batch of graphs or a single large graph.
-      conv_receivers = jnp.concatenate((receivers, jnp.arange(total_num_nodes)),
+        # We add self edges to the senders and receivers so that each node
+        # includes itself in aggregation.
+        # In principle, a `GraphsTuple` should partition by n_edge, but in
+        # this case it is not required since a GCN is agnostic to whether
+        # the `GraphsTuple` is a batch of graphs or a single large graph.
+        conv_receivers = jnp.concatenate((receivers, jnp.arange(total_num_nodes)),
                                        axis=0)
-      conv_senders = jnp.concatenate((senders, jnp.arange(total_num_nodes)),
+        conv_senders = jnp.concatenate((senders, jnp.arange(total_num_nodes)),
                                      axis=0)
     else:
-      conv_senders = senders
-      conv_receivers = receivers
+        conv_senders = senders
+        conv_receivers = receivers
 
     # pylint: disable=g-long-lambda
     if symmetric_normalization:
@@ -594,3 +595,17 @@ def _ApplyGCN(graph):
     return graph._replace(nodes=nodes)
 
   return _ApplyGCN
+
+
+def random_graph(device=None):
+    """Returns a random graph with 10 nodes and 20 edges.
+    
+    Args:
+        device: Optional device to place the arrays on. If None, uses current device.
+    """
+    n_node = 10
+    n_edge = 20
+    with jax.device(device) if device else nullcontext():
+        senders = jnp.random.randint(0, n_node, size=n_edge)
+        receivers = jnp.random.randint(0, n_node, size=n_edge)
+        # ...