diff --git a/docs/source/nim.md b/docs/source/nim.md
index d479aedcd..862a88c83 100644
--- a/docs/source/nim.md
+++ b/docs/source/nim.md
@@ -9,10 +9,10 @@ Thus, an application can call the model without loading its weights or managing
 NIMs for structured data models follow SDM's unified prediction interface.
 We have published the following NIMs:
 
-| Model ID          | Model                                  | Deployment Instructions                                                            |
-| ----------------- | -------------------------------------- | ---------------------------------------------------------------------------------- |
-| `kumo-tabular`    | {py:class}`~sdm.models.KumoTabular`    | [Kumo Tabular on NVIDIA Build](https://build.nvidia.com/nvidia/kumo-tabular)       |
-| `kumo-relational` | {py:class}`~sdm.models.KumoRelational` | [Kumo Relational on NVIDIA Build](https://build.nvidia.com/nvidia/kumo-relational) |
+| Model ID          | Model                                  | NVIDIA Build                                              | NGC Catalog                                                                              |
+| ----------------- | -------------------------------------- | --------------------------------------------------------- | ---------------------------------------------------------------------------------------- |
+| `kumo-tabular`    | {py:class}`~sdm.models.KumoTabular`    | [Deploy](https://build.nvidia.com/nvidia/kumo-tabular)    | [Container](https://catalog.ngc.nvidia.com/orgs/nim/nvidia/containers/kumo-tabular/-)    |
+| `kumo-relational` | {py:class}`~sdm.models.KumoRelational` | [Deploy](https://build.nvidia.com/nvidia/kumo-relational) | [Container](https://catalog.ngc.nvidia.com/orgs/nim/nvidia/containers/kumo-relational/-) |
 
 ## The NIM API