diff --git a/pkg/apis/ome/v1beta1/model.go b/pkg/apis/ome/v1beta1/model.go
index d16a78aba..8dfc197b0 100644
--- a/pkg/apis/ome/v1beta1/model.go
+++ b/pkg/apis/ome/v1beta1/model.go
@@ -286,7 +286,7 @@ const (
 // ModelCapability enum
 // TODO: Remove legacy capabilities
 //
-// +kubebuilder:validation:Enum=TEXT_GENERATION;TEXT_SUMMARIZATION;TEXT_EMBEDDINGS;TEXT_RERANK;CHAT;VISION;EMBEDDING;RERANK;TEXT_TO_TEXT;TEXT_TO_AUDIO;TEXT_TO_IMAGE;TEXT_TO_VIDEO;IMAGE_TEXT_TO_TEXT;IMAGE_TEXT_TO_AUDIO;IMAGE_TEXT_TO_IMAGE;IMAGE_TEXT_TO_VIDEO;VIDEO_TEXT_TO_AUDIO;AUDIO_TO_TEXT;AUDIO_TO_AUDIO;AUDIO_TEXT_TO_TEXT;AUDIO_TRANSLATION
+// +kubebuilder:validation:Enum=TEXT_GENERATION;TEXT_SUMMARIZATION;TEXT_EMBEDDINGS;TEXT_RERANK;CHAT;VISION;EMBEDDING;RERANK;TEXT_TO_TEXT;TEXT_TO_AUDIO;TEXT_TO_IMAGE;TEXT_TO_VIDEO;IMAGE_TEXT_TO_TEXT;IMAGE_TEXT_TO_AUDIO;IMAGE_TEXT_TO_IMAGE;IMAGE_TEXT_TO_VIDEO;VIDEO_TEXT_TO_AUDIO;VIDEO_TEXT_TO_TEXT;AUDIO_TO_TEXT;AUDIO_TO_AUDIO;AUDIO_TEXT_TO_TEXT;AUDIO_TRANSLATION
 type ModelCapability string
 
 const (
@@ -310,6 +310,7 @@ const (
 	ModelCapabilityImageTextToImage ModelCapability = "IMAGE_TEXT_TO_IMAGE"
 	ModelCapabilityImageTextToVideo ModelCapability = "IMAGE_TEXT_TO_VIDEO"
 	ModelCapabilityVideoTextToAudio ModelCapability = "VIDEO_TEXT_TO_AUDIO"
+	ModelCapabilityVideoTextToText  ModelCapability = "VIDEO_TEXT_TO_TEXT"
 	ModelCapabilityAudioToText      ModelCapability = "AUDIO_TO_TEXT"
 	ModelCapabilityAudioToAudio     ModelCapability = "AUDIO_TO_AUDIO"
 	ModelCapabilityAudioTextToText  ModelCapability = "AUDIO_TEXT_TO_TEXT"
diff --git a/pkg/modelconfig/capability.go b/pkg/modelconfig/capability.go
index bf0bdff39..ca72ecf5b 100644
--- a/pkg/modelconfig/capability.go
+++ b/pkg/modelconfig/capability.go
@@ -21,6 +21,7 @@ const (
 	CapabilityTextToAudio      Capability = "TEXT_TO_AUDIO"
 	CapabilityImageTextToAudio Capability = "IMAGE_TEXT_TO_AUDIO"
 	CapabilityVideoTextToAudio Capability = "VIDEO_TEXT_TO_AUDIO"
+	CapabilityVideoTextToText  Capability = "VIDEO_TEXT_TO_TEXT"
 	CapabilityAudioToText      Capability = "AUDIO_TO_TEXT"
 	CapabilityAudioToAudio     Capability = "AUDIO_TO_AUDIO"
 	CapabilityAudioTextToText  Capability = "AUDIO_TEXT_TO_TEXT"
@@ -166,14 +167,21 @@ func diffusionRule(hf HuggingFaceModel) []Capability {
 }
 
 func nemotronHNanoRule(hf HuggingFaceModel) []Capability {
-	if !strings.Contains(strings.ToLower(hf.GetModelType()), "nemotronh_nano") {
+	modelType := strings.ToLower(hf.GetModelType())
+	if !strings.Contains(modelType, "nemotronh_nano") {
 		return nil
 	}
-	return []Capability{
+	capabilities := []Capability{
 		CapabilityImageTextToText,
 		CapabilityTextToText,
-		CapabilityAudioToText,
 	}
+	if strings.Contains(modelType, "nemotronh_nano_omni") {
+		capabilities = append(capabilities,
+			CapabilityAudioTextToText,
+			CapabilityVideoTextToText,
+		)
+	}
+	return capabilities
 }
 
 func omniRule(hf HuggingFaceModel) []Capability {
diff --git a/pkg/modelconfig/capability_test.go b/pkg/modelconfig/capability_test.go
index 34c20839d..9d6881fd5 100644
--- a/pkg/modelconfig/capability_test.go
+++ b/pkg/modelconfig/capability_test.go
@@ -198,7 +198,23 @@ func TestClassifyCapabilities_NemotronH_Nano(t *testing.T) {
 	want := []Capability{
 		CapabilityImageTextToText,
 		CapabilityTextToText,
-		CapabilityAudioToText,
+		CapabilityAudioTextToText,
+		CapabilityVideoTextToText,
+	}
+	if got := classifyCapabilities(m); !equalCaps(got, want) {
+		t.Errorf("got %v, want %v", got, want)
+	}
+}
+
+func TestClassifyCapabilities_NemotronH_NanoNonOmni(t *testing.T) {
+	m := &stubModel{
+		modelType:    "NemotronH_Nano_VL_V2",
+		architecture: "NemotronH_Nano_VL_V2",
+		hasVision:    true,
+	}
+	want := []Capability{
+		CapabilityImageTextToText,
+		CapabilityTextToText,
 	}
 	if got := classifyCapabilities(m); !equalCaps(got, want) {
 		t.Errorf("got %v, want %v", got, want)
@@ -413,7 +429,8 @@ func TestClassifyCapabilities_RegressionSet(t *testing.T) {
 			want: []Capability{
 				CapabilityImageTextToText,
 				CapabilityTextToText,
-				CapabilityAudioToText,
+				CapabilityAudioTextToText,
+				CapabilityVideoTextToText,
 			},
 		},
 		{
diff --git a/pkg/modelparser/capability.go b/pkg/modelparser/capability.go
index 2ed774599..98b2c9ac0 100644
--- a/pkg/modelparser/capability.go
+++ b/pkg/modelparser/capability.go
@@ -23,6 +23,7 @@ var capabilityToOME = map[modelconfig.Capability]v1beta1.ModelCapability{
 	modelconfig.CapabilityTextToAudio:      v1beta1.ModelCapabilityTextToAudio,
 	modelconfig.CapabilityImageTextToAudio: v1beta1.ModelCapabilityImageTextToAudio,
 	modelconfig.CapabilityVideoTextToAudio: v1beta1.ModelCapabilityVideoTextToAudio,
+	modelconfig.CapabilityVideoTextToText:  v1beta1.ModelCapabilityVideoTextToText,
 	modelconfig.CapabilityAudioToText:      v1beta1.ModelCapabilityAudioToText,
 	modelconfig.CapabilityAudioToAudio:     v1beta1.ModelCapabilityAudioToAudio,
 	modelconfig.CapabilityAudioTextToText:  v1beta1.ModelCapabilityAudioTextToText,
diff --git a/pkg/modelparser/capability_test.go b/pkg/modelparser/capability_test.go
index 8b1eb64ef..250b1e0dd 100644
--- a/pkg/modelparser/capability_test.go
+++ b/pkg/modelparser/capability_test.go
@@ -29,6 +29,7 @@ func TestCapabilityToOME(t *testing.T) {
 		{modelconfig.CapabilityTextToAudio, v1beta1.ModelCapabilityTextToAudio},
 		{modelconfig.CapabilityImageTextToAudio, v1beta1.ModelCapabilityImageTextToAudio},
 		{modelconfig.CapabilityVideoTextToAudio, v1beta1.ModelCapabilityVideoTextToAudio},
+		{modelconfig.CapabilityVideoTextToText, v1beta1.ModelCapabilityVideoTextToText},
 		{modelconfig.CapabilityAudioToText, v1beta1.ModelCapabilityAudioToText},
 		{modelconfig.CapabilityAudioToAudio, v1beta1.ModelCapabilityAudioToAudio},
 		{modelconfig.CapabilityAudioTextToText, v1beta1.ModelCapabilityAudioTextToText},