diff --git a/docs/clearml_agent/clearml_agent_nvcr.md b/docs/clearml_agent/clearml_agent_nvcr.md
new file mode 100644
index 00000000..3bdafbdb
--- /dev/null
+++ b/docs/clearml_agent/clearml_agent_nvcr.md
@@ -0,0 +1,43 @@
+---
+title: NVCR Access
+---
+
+To allow ClearML Agents to access NVIDIA's container registry (`nvcr.io`), the machine’s Docker infrastructure must first be configured with valid NGC credentials.
+This enables Agents to pull NVIDIA-provided containers, such as those used by the [NVIDIA NIM app](../webapp/applications/apps_nvidia_nim.md). The setup is 
+required once per worker node, not every time you run an app.
+
+Configure Docker access to the `nvcr` repository on [bare-metal/VM](#on-bare-metal--vm) or [Kubernetes](#on-kubernetes). 
+
+## On Bare Metal / VM
+   
+Execute the following command where the agent that will execute the app instance will be running (replace the password with a valid NGC API key):
+
+```
+docker login nvcr.io --username '$oauthtoken' --password 'nvapi-**'
+```
+Password is provided with your `nvcr` account.
+
+## On Kubernetes
+  
+To make `nvcr` available to agents running on Kubernetes:
+* Create an `nvcr-registry` secret in the same namespace where the agent is running. Replace:
+  * `<NAMESPACE>` with the namespace where your ClearML Agent is deployed
+  * `<USERNAME>` with your NVIDIA registry username
+  * `<PASSWORD>` with your valid NGC API key <br/><br/>
+    
+  ```
+  kubectl create secret docker-registry nvcr-registry -n <NAMESPACE> \
+    --docker-server=nvcr.io \
+    --docker-username=<USERNAME> \
+    --docker-password=<PASSWORD> \
+    --docker-email=""
+  ```
+   
+* Configure image pull secrets for the NVIDIA registry.
+  In your Agent Helm values override, add:
+
+  ```
+  imageCredentials:
+    extraImagePullSecrets:
+      - name: nvcr-registry
+  ```
diff --git a/docs/webapp/applications/apps_nvidia_nim.md b/docs/webapp/applications/apps_nvidia_nim.md
index 8fb0e928..796f198d 100644
--- a/docs/webapp/applications/apps_nvidia_nim.md
+++ b/docs/webapp/applications/apps_nvidia_nim.md
@@ -11,10 +11,9 @@ serves your model on a machine of your choice. Once an app instance is running,
 publicly accessible network endpoint. The app monitors endpoint activity and shuts down if the model remains inactive 
 for a specified maximum idle time.
 
-Note that the `NGC_API_KEY` environment variable needs to be set to a valid NGC API key. You can set the variable in one 
-of the following ways:
-* The NIM app deployment form’s `Environment Variables` field
-* [Configuration vault](../settings/webapp_settings_profile.md#configuration-vault)
+* The `NGC_API_KEY` environment variable needs to be set to a valid NGC API key. You can set the variable in one of the following ways:
+  * The NIM app deployment form’s `Environment Variables` field
+  * [Configuration vault](../settings/webapp_settings_profile.md#configuration-vault)
 
 :::info AI Application Gateway
 The NIM app makes use of the App Gateway Router which implements a secure, authenticated network endpoint for the model.
@@ -81,7 +80,8 @@ values from the file, which can be modified before launching the app instance
 * **Application instance project**: The ClearML project where the app instance is created. Access is determined by 
   project-level permissions (i.e. users with read access can use the app).
 * **NIM Container Image**: Select the containerized application image to use. Note the different tags / versions of each image
-* **Compute Resource (Queue)**: The [ClearML Queue](../../fundamentals/agents_and_queues.md#what-is-a-queue)  to which the NIM app instance task will be enqueued (make sure an agent is assigned to it)
+* **Compute Resource (Queue)**: The [ClearML Queue](../../fundamentals/agents_and_queues.md#what-is-a-queue)  to which the NIM app instance task will be enqueued. Make sure an agent 
+  is assigned to this queue and has access to NVIDIA's container registry (`nvcr.io`). See [NVCR Access](../../clearml_agent/clearml_agent_nvcr.md) for more information.
 * **AI Gateway Route**: Select an available, admin-preconfigured route to use as the service endpoint. If none is selected, an ephemeral endpoint will be created.
 * **Idle Time Limit** (Hours): Maximum idle time after which the app instance will shut down
 * **Environment Variables**: Additional environment variable to set inside the container before launching the application 
diff --git a/sidebars.js b/sidebars.js
index a26f9f67..723430fd 100644
--- a/sidebars.js
+++ b/sidebars.js
@@ -668,6 +668,7 @@ module.exports = {
                     ]
                 },
                 'clearml_agent/multi_node_training',
+                'clearml_agent/clearml_agent_nvcr',
                 {
                     type: 'category',
                     collapsible: true,