feat: update GAIE to release version with hints in headers (#5503)

Signed-off-by: Anna Tchernych <atchernych@nvidia.com>

deploy/inference-gateway/helm/dynamo-gaie/templates/inference-pool.yaml

@@ -12,6 +12,30 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+
+# InferencePool for kGateway (stable API - inference.networking.k8s.io/v1)
+# This is used by kGateway to resolve HTTPRoute backends
+apiVersion: inference.networking.k8s.io/v1
+kind: InferencePool
+metadata:
+  name: {{ .Values.model.shortName }}-pool
+  namespace: {{ .Release.Namespace }}
+spec:
+  targetPorts:
+    - number: {{ .Values.inferencePool.port }}
+  selector:
+    matchLabels:
+      nvidia.com/dynamo-component: Frontend
+      nvidia.com/dynamo-namespace: {{ include "dynamo-gaie.dynamoNamespace" . }}
+  endpointPickerRef:
+    group: ""
+    kind: Service
+    name: {{ .Values.model.shortName }}-epp
+    port:
+      number: 9002
+---
+# InferencePool for EPP (experimental API - inference.networking.x-k8s.io/v1alpha2)
+# This is used by the EPP to discover and route to backend pods
 apiVersion: inference.networking.x-k8s.io/v1alpha2
 kind: InferencePool
 metadata:

deploy/inference-gateway/helm/dynamo-gaie/values.yaml

@@ -27,8 +27,6 @@ model:
   identifier: "Qwen/Qwen3-0.6B"
   # This is the short name of the model that will be used to generate the resource names
   shortName: "qwen"
-  # Criticality level for the inference model
-  criticality: "Critical"
 
 # InferencePool configuration
 inferencePool:
@@ -41,6 +39,9 @@ httpRoute:
   enabled: true
   # Gateway parent reference configuration
   gatewayName: "inference-gateway"
+  # Namespace where the Gateway resource is deployed
+  # Leave empty ("") to use the release namespace, or set explicitly (e.g., "default")
+  gatewayNamespace: ""
   # Path matching configuration
   path:
     prefix: "/"
@@ -51,8 +52,8 @@ httpRoute:
 extension:
   # EPP image for the GAIE extension (Dynamo EPP image by default)
   image: "" # leave empty to use defaults below
-  standardImage: us-central1-docker.pkg.dev/k8s-artifacts-prod/images/gateway-api-inference-extension/epp:v0.4.0
-  dynamoImage: nvcr.io/nvstaging/ai-dynamo/gaie-epp-dynamo:v0.6.0-1
+  standardImage: us-central1-docker.pkg.dev/k8s-artifacts-prod/images/gateway-api-inference-extension/epp:v1.2.1
+  dynamoImage: gitlab-master.nvidia.com:5005/dl/ai-dynamo/dynamo/epp-inference-extension-dynamo:new-build-1
 
 # generic knobs you may want in both modes
 imagePullSecrets:

deploy/inference-gateway/install_gaie_crd_kgateway.sh → deploy/inference-gateway/scripts/install_gaie_crd_kgateway.sh

@@ -18,29 +18,26 @@
 set -euo pipefail
 trap 'echo "Error at line $LINENO. Exiting."' ERR
 
-
-MODEL_NAMESPACE=my-model
-kubectl create namespace $MODEL_NAMESPACE || true
-
 # Install the Gateway API
-GATEWAY_API_VERSION=v1.3.0
+GATEWAY_API_VERSION=v1.4.1
 kubectl apply -f https://github.com/kubernetes-sigs/gateway-api/releases/download/$GATEWAY_API_VERSION/standard-install.yaml
 
 
 # Install the Inference Extension CRDs
-INFERENCE_EXTENSION_VERSION=v0.5.1
-kubectl apply -f https://github.com/kubernetes-sigs/gateway-api-inference-extension/releases/download/$INFERENCE_EXTENSION_VERSION/manifests.yaml -n  $MODEL_NAMESPACE
+IGW_LATEST_RELEASE=v1.2.1
+kubectl apply -f https://github.com/kubernetes-sigs/gateway-api-inference-extension/releases/download/${IGW_LATEST_RELEASE}/manifests.yaml
 
 
 # Install and upgrade Kgateway (includes CRDs)
-KGATEWAY_VERSION=v2.0.3
-KGATEWAY_SYSTEM_NAMESPACE=kgateway-system
-kubectl create namespace $KGATEWAY_SYSTEM_NAMESPACE || true
-
-helm upgrade -i --create-namespace --namespace $KGATEWAY_SYSTEM_NAMESPACE --version $KGATEWAY_VERSION kgateway-crds oci://cr.kgateway.dev/kgateway-dev/charts/kgateway-crds
+KGTW_VERSION=v2.1.1
+helm upgrade -i --create-namespace --namespace kgateway-system --version $KGTW_VERSION \
+  kgateway-crds oci://cr.kgateway.dev/kgateway-dev/charts/kgateway-crds
 
-helm upgrade -i --namespace $KGATEWAY_SYSTEM_NAMESPACE --version $KGATEWAY_VERSION kgateway oci://cr.kgateway.dev/kgateway-dev/charts/kgateway --set inferenceExtension.enabled=true
+helm upgrade -i --namespace kgateway-system --version $KGTW_VERSION kgateway \
+  oci://cr.kgateway.dev/kgateway-dev/charts/kgateway \
+  --set inferenceExtension.enabled=true
 
+kubectl apply -f https://raw.githubusercontent.com/kubernetes-sigs/gateway-api-inference-extension/refs/tags/${IGW_LATEST_RELEASE}/config/manifests/gateway/kgateway/gateway.yaml
 
-# Deploy the Gateway Instance
-kubectl apply -f https://github.com/kubernetes-sigs/gateway-api-inference-extension/raw/v1.0.0/config/manifests/gateway/kgateway/gateway.yaml -n $MODEL_NAMESPACE
+kubectl patch gateway inference-gateway --type='json' \
+  -p='[{"op": "replace", "path": "/spec/gatewayClassName", "value": "kgateway"}]'