diff --git a/tests/python/unittest/test_custom_datatypes_change_dtype.py b/tests/python/unittest/test_custom_datatypes_change_dtype.py
index a13f912141da..1615f927171e 100644
--- a/tests/python/unittest/test_custom_datatypes_change_dtype.py
+++ b/tests/python/unittest/test_custom_datatypes_change_dtype.py
@@ -20,6 +20,7 @@
 import numpy as np
 from tvm import relay
 from tvm.relay.testing.inception_v3 import get_workload as get_inception
+from tvm.relay.testing.resnet import get_workload as get_resnet
 
 tgt = "llvm"
 
@@ -92,8 +93,29 @@ def test_change_dtype_simple():
     result_converted = convert_ndarray('float32', result, ex)
     print(result_converted)
 
-def test_change_dtype_inception_v3():
+def test_change_dtype_resnet():
+    expr, params = get_resnet()
+
+    ex = relay.create_executor("graph")
+
+    src_dtype = 'float32'
+    dst_dtype = 'custom[bfloat]16' # Change me to posit.
+    expr, params = change_dtype(src_dtype, dst_dtype, expr, params)
+
+    # Convert the input into the correct format.
+    input = tvm.nd.array(np.random.rand(3, 299, 299).astype(src_dtype))
+    input = convert_ndarray(dst_dtype, input, ex)
 
+    def print_info(node):
+        if not isinstance(node, relay.op.op.Op):
+            if ("custom[bfloat]32" not in str(node.checked_type())):
+                print(node.checked_type())
+    relay.ir_pass.post_order_visit(expr, print_info)
+
+    # Execute the model in the new datatype.
+    result = ex.evaluate(expr)(input, **params)
+
+def test_change_dtype_inception_v3():
     expr, params = get_inception()
 
     ex = relay.create_executor("graph")
@@ -120,3 +142,4 @@ def print_info(node):
     setup()
     # test_change_dtype_inception_v3()
     test_change_dtype_simple()
+    test_change_dtype_resnet()