diff --git a/_topic_map.yml b/_topic_map.yml
index 234aa6c4b6..43083aebce 100644
--- a/_topic_map.yml
+++ b/_topic_map.yml
@@ -1704,11 +1704,6 @@ Topics:
 # Apps
 - Name: Creating and managing serverless applications
   File: serving-creating-managing-apps
-# HA
-- Name: High availability on OpenShift Serverless
-  File: serverless-HA
-- Name: Tracing requests
-  File: serverless-tracing
 # Knative CLI
 - Name: Knative CLI
   Dir: knative_cli
@@ -1752,6 +1747,18 @@ Topics:
 # Channels
   - Name: Using channels
     File: serverless-channels
+# HA
+- Name: High availability on OpenShift Serverless
+  File: serverless-HA
+# Tracing
+- Name: Tracing requests
+  File: serverless-tracing
 # Metering
 - Name: Using metering with OpenShift Serverless
   File: serverless-metering
+# Integrations
+# - Name: Integrations
+#  Dir: integrations
+#  Topics:
+#  - Name: Using NVIDIA GPU resources with serverless applications
+#    File: gpu-resources
diff --git a/serverless/integrations/gpu-resources.adoc b/serverless/integrations/gpu-resources.adoc
new file mode 100644
index 0000000000..8d7ca7de71
--- /dev/null
+++ b/serverless/integrations/gpu-resources.adoc
@@ -0,0 +1,39 @@
+include::modules/serverless-document-attributes.adoc[]
+[id="gpu-resources"]
+= Using NVIDIA GPU resources with serverless applications
+:context: gpu-resources
+include::modules/common-attributes.adoc[]
+
+toc::[]
+
+NVIDIA supports experimental use of GPU resources on {product-title}.
+See link:https://docs.nvidia.com/datacenter/kubernetes/openshift-on-gpu-install-guide/index.html[{product-title} on NVIDIA GPU accelerated clusters] for more information about setting up GPU resources on {product-title}.
+
+After GPU resources are enabled for your {product-title} cluster, you can specify GPU requirements for a Knative service using the `kn` CLI.
+
+.Procedure
+
+You can specify a GPU resource requirement when you create a Knative service using `kn`.
+
+. Create a service.
+. Set the GPU resource requirement limit to `1` by using `nvidia.com/gpu=1`:
++
+----
+$ kn service create hello --image docker.io/knativesamples/hellocuda-go --limit nvidia.com/gpu=1
+----
++
+A GPU resource requirement limit of `1` means that the service has 1 GPU resource dedicated.
+Services do not share GPU resources. Any other services that require GPU resources must wait until the GPU resource is no longer in use.
++
+A limit of 1 GPU also means that applications exceeding usage of 1 GPU resource are restricted.
+If a service requests more than 1 GPU resource, it is deployed on a node where the GPU resource requirements can be met.
+
+.Updating GPU requirements for a Knative service using `kn`
+
+* Update the service. Change the GPU resource requirement limit to `3` by using `nvidia.com/gpu=3`:
+----
+$ kn service update hello --limit nvidia.com/gpu=3
+----
+
+== Additional resources
+* For more information about limits, see xref:../../applications/quotas/quotas-setting-per-project.adoc[Setting resource quotas for extended resources].
diff --git a/serverless/integrations/images b/serverless/integrations/images
new file mode 120000
index 0000000000..5e67573196
--- /dev/null
+++ b/serverless/integrations/images
@@ -0,0 +1 @@
+../images
\ No newline at end of file
diff --git a/serverless/integrations/modules b/serverless/integrations/modules
new file mode 120000
index 0000000000..464b823aca
--- /dev/null
+++ b/serverless/integrations/modules
@@ -0,0 +1 @@
+../modules
\ No newline at end of file