diff --git a/hermes_cli/models_catalog_static.py b/hermes_cli/models_catalog_static.py
index 62be10e8521c0..7617964ec0a98 100644
--- a/hermes_cli/models_catalog_static.py
+++ b/hermes_cli/models_catalog_static.py
@@ -199,7 +199,7 @@ def _xai_curated_models() -> list[str]:
     "xai": list(_XAI_MODELS),
     # Nemotron flagships, then third-party agentic models hosted on build.nvidia.com.
     "nvidia": [
-        "nvidia/nemotron-3.5-super-vl-120b-a12b", "nvidia/nemotron-3-ultra-550b-a55b", "nvidia/nemotron-3-super-120b-a12b",
+        "nvidia/nemotron-3.5-super-vl-120b-a12b", "nvidia/nemotron-3-ultra-550b-a55b",
         "nvidia/nemotron-3.5-lightning-30b-a3b", "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning",
         "z-ai/glm-5.3", "z-ai/glm-5.2", "moonshotai/kimi-k2.6", "minimaxai/minimax-m3",
     ],
diff --git a/tests/hermes_cli/test_models_dev_preferred_merge.py b/tests/hermes_cli/test_models_dev_preferred_merge.py
index 96660215d12bd..dc089fa1fd6d1 100644
--- a/tests/hermes_cli/test_models_dev_preferred_merge.py
+++ b/tests/hermes_cli/test_models_dev_preferred_merge.py
@@ -59,6 +59,11 @@ def test_merge_case_insensitive_dedup(self):
 
 class TestProviderModelIdsPreferred:
 
+    def test_nvidia_curated_models_replace_retired_super(self):
+        nvidia_models = _PROVIDER_MODELS["nvidia"]
+        assert nvidia_models[0] == "nvidia/nemotron-3.5-super-vl-120b-a12b"
+        assert "nvidia/nemotron-3-super-120b-a12b" not in nvidia_models
+
     def test_k3_live_discovery_is_scoped_to_kimi_coding_endpoint(self):
         """Coding keys discover K3; legacy Moonshot keys must not advertise it."""
 
diff --git a/website/docs/integrations/providers.md b/website/docs/integrations/providers.md
index 7e85f0bea90d1..b0d814872a049 100644
--- a/website/docs/integrations/providers.md
+++ b/website/docs/integrations/providers.md
@@ -567,7 +567,7 @@ Supported models: `MiniMax-M2.7` (main) and `MiniMax-M2.7-highspeed` (wired as t
 
 Use the NVIDIA NIM provider to access Nemotron and other open-source models through NVIDIA’s hosted service or a local NIM endpoint. Model identifiers for hosted endpoints and instructions for local deployment are available at [build.nvidia.com/models](https://build.nvidia.com/models).
 
-:::note The examples below use Nemotron 3.5 Super VL, which supports text and image inputs, reasoning, and tool calls.
+:::note The examples below use Nemotron 3.5 Super VL, which supports text, image, and video inputs, reasoning, and tool calls.
 
 #### Cloud (build.nvidia.com)