ray-project · jjyao · Nov 7, 2025 · Nov 7, 2025 · Nov 7, 2025 · Nov 7, 2025
@@ -19,7 +19,7 @@ Note that the YAML file in this example uses `serveConfigV2`. You need KubeRay v
 
 ```sh
 # Create a RayService
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-service.mobilenet.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-service.mobilenet.yaml
 ```
 
 * The [mobilenet.py](https://github.com/ray-project/serve_config_examples/blob/master/mobilenet/mobilenet.py) file needs `tensorflow` as a dependency. Hence, the YAML file uses `rayproject/ray-ml` image instead of `rayproject/ray` image.

@@ -37,12 +37,12 @@ The KubeRay operator Pod must be on the CPU node if you have set up the taint fo
 
 ## Step 2: Submit the RayJob
 
-Create the RayJob custom resource with [ray-job.batch-inference.yaml](https://github.com/ray-project/kuberay/blob/v1.4.2/ray-operator/config/samples/ray-job.batch-inference.yaml).
+Create the RayJob custom resource with [ray-job.batch-inference.yaml](https://github.com/ray-project/kuberay/blob/v1.5.0/ray-operator/config/samples/ray-job.batch-inference.yaml).
 
 Download the file with `curl`:
 
 ```bash
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-job.batch-inference.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-job.batch-inference.yaml
 ```
 
 Note that the `RayJob` spec contains a spec for the `RayCluster`. This tutorial uses a single-node cluster with 4 GPUs. For production use cases, use a multi-node cluster where the head node doesn't have GPUs, so that Ray can automatically schedule GPU workloads on worker nodes which won't interfere with critical Ray processes on the head node.

@@ -17,15 +17,15 @@ kind create cluster --image=kindest/node:v1.26.0
 ```sh
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-# Install both CRDs and KubeRay operator v1.4.2.
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2
+# Install both CRDs and KubeRay operator v1.5.0.
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0
 ```
 
 ### Method 2: Kustomize
 
 ```sh
 # Install CRD and KubeRay operator.
-kubectl create -k "github.com/ray-project/kuberay/ray-operator/config/default?ref=v1.4.2"
+kubectl create -k "github.com/ray-project/kuberay/ray-operator/config/default?ref=v1.5.0"
 ```
 
 ## Step 3: Validate Installation

@@ -28,7 +28,7 @@ Once the KubeRay operator is running, you're ready to deploy a RayCluster. Creat
 
 ```sh
 # Deploy a sample RayCluster CR from the KubeRay Helm chart repo:
-helm install raycluster kuberay/ray-cluster --version 1.4.2
+helm install raycluster kuberay/ray-cluster --version 1.5.0
 ```
 
 

@@ -96,7 +96,7 @@ Follow the [KubeRay Operator Installation](kuberay-operator-deploy) to install t
 ## Step 3: Install a RayJob
 
 ```sh
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-job.sample.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-job.sample.yaml
 ```
 
 ## Step 4: Verify the Kubernetes cluster status
@@ -163,13 +163,13 @@ The Python script `sample_code.py` used by `entrypoint` is a simple Ray script t
 ## Step 6: Delete the RayJob
 
 ```sh
-kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-job.sample.yaml
+kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-job.sample.yaml
 ```
 
 ## Step 7: Create a RayJob with `shutdownAfterJobFinishes` set to true
 
 ```sh
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-job.shutdown.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-job.shutdown.yaml
 ```
 
 The `ray-job.shutdown.yaml` defines a RayJob custom resource with `shutdownAfterJobFinishes: true` and `ttlSecondsAfterFinished: 10`.
@@ -197,7 +197,7 @@ kubectl get raycluster
 
 ```sh
 # Step 10.1: Delete the RayJob
-kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-job.shutdown.yaml
+kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-job.shutdown.yaml
 
 # Step 10.2: Delete the KubeRay operator
 helm uninstall kuberay-operator

@@ -3,7 +3,7 @@
 
 ## Prerequisites
 
-This guide mainly focuses on the behavior of KubeRay v1.4.2 and Ray 2.46.0.
+This guide mainly focuses on the behavior of KubeRay v1.5.0 and Ray 2.46.0.
 
 ## What's a RayService?
 
@@ -35,7 +35,7 @@ Note that the YAML file in this example uses `serveConfigV2` to specify a multi-
 ## Step 3: Install a RayService
 
 ```sh
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-service.sample.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-service.sample.yaml
 ```
 
 ## Step 4: Verify the Kubernetes cluster status
@@ -129,7 +129,7 @@ curl -X POST -H 'Content-Type: application/json' rayservice-sample-serve-svc:800
 
 ```sh
 # Delete the RayService.
-kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-service.sample.yaml
+kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-service.sample.yaml
 
 # Uninstall the KubeRay operator.
 helm uninstall kuberay-operator

@@ -33,10 +33,10 @@ Four examples show how to use ingress to access your Ray cluster:
 # Step 1: Install KubeRay operator and CRD
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0
 
 # Step 2: Install a RayCluster
-helm install raycluster kuberay/ray-cluster --version 1.4.2
+helm install raycluster kuberay/ray-cluster --version 1.5.0
 
 # Step 3: Edit the `ray-operator/config/samples/ray-cluster-alb-ingress.yaml`
 #
@@ -123,10 +123,10 @@ Now run the following commands:
 # Step 1: Install KubeRay operator and CRD
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0
 
 # Step 2: Install a RayCluster
-helm install raycluster kuberay/ray-cluster --version 1.4.2
+helm install raycluster kuberay/ray-cluster --version 1.5.0
 
 # Step 3: Edit ray-cluster-gclb-ingress.yaml to replace the service name with the name of the head service from the RayCluster. (Output of `kubectl get svc`)
 
@@ -186,12 +186,12 @@ kubectl wait --namespace ingress-nginx \
 # Step 3: Install KubeRay operator and CRD
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0
 
 # Step 4: Install RayCluster and create an ingress separately.
 # More information about change of setting was documented in https://github.com/ray-project/kuberay/pull/699
 # and `ray-operator/config/samples/ray-cluster.separate-ingress.yaml`
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-cluster.separate-ingress.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-cluster.separate-ingress.yaml
 kubectl apply -f ray-cluster.separate-ingress.yaml
 
 # Step 5: Check the ingress created in Step 4.
@@ -230,10 +230,10 @@ kubectl describe ingress raycluster-ingress-head-ingress
 # Step 1: Install KubeRay operator and CRD
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0
 
 # Step 2: Install a RayCluster
-helm install raycluster kuberay/ray-cluster --version 1.4.2
+helm install raycluster kuberay/ray-cluster --version 1.5.0
 
 # Step 3: Edit the `ray-operator/config/samples/ray-cluster-agc-gatewayapi.yaml`
 #

@@ -66,7 +66,7 @@ In this mode, you _must_ disable the KubeRay init container injection by setting
 
 ```bash
 # Set ENABLE_INIT_CONTAINER_INJECTION=false on the KubeRay operator.
-helm upgrade kuberay-operator kuberay/kuberay-operator --version 1.4.2 \
+helm upgrade kuberay-operator kuberay/kuberay-operator --version 1.5.0 \
   --set env\[0\].name=ENABLE_INIT_CONTAINER_INJECTION \
   --set-string env\[0\].value=false
 

@@ -56,7 +56,7 @@ kubectl get all -n prometheus-system
 * Set `metrics.serviceMonitor.enabled=true` when installing the KubeRay operator with Helm to create a ServiceMonitor that scrapes metrics exposed by the KubeRay operator's service.
   ```sh
   # Enable the ServiceMonitor and set the label `release: prometheus` to the ServiceMonitor so that Prometheus can discover it
-  helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2 \
+  helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0 \
     --set metrics.serviceMonitor.enabled=true \ 
     --set metrics.serviceMonitor.selector.release=prometheus 
   ```
@@ -104,7 +104,7 @@ curl localhost:8080
   * `# HELP`: Describe the meaning of this metric.
   * `# TYPE`: See [this document](https://prometheus.io/docs/concepts/metric_types/) for more details.
 
-* Three required environment variables are defined in [ray-cluster.embed-grafana.yaml](https://github.com/ray-project/kuberay/blob/v1.4.2/ray-operator/config/samples/ray-cluster.embed-grafana.yaml). See [Configuring and Managing Ray Dashboard](https://docs.ray.io/en/latest/cluster/configure-manage-dashboard.html) for more details about these environment variables.
+* Three required environment variables are defined in [ray-cluster.embed-grafana.yaml](https://github.com/ray-project/kuberay/blob/v1.5.0/ray-operator/config/samples/ray-cluster.embed-grafana.yaml). See [Configuring and Managing Ray Dashboard](https://docs.ray.io/en/latest/cluster/configure-manage-dashboard.html) for more details about these environment variables.
   ```yaml
   env:
     - name: RAY_GRAFANA_IFRAME_HOST

@@ -29,7 +29,7 @@ You need to have the access to configure Kubernetes control plane to replace the
 KubeRay v1.4.0 and later versions support scheduler plugins.
 
 ```sh
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2 --set batchScheduler.name=scheduler-plugins
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0 --set batchScheduler.name=scheduler-plugins
 ```
 
 ## Step 4: Deploy a RayCluster with gang scheduling

@@ -35,7 +35,7 @@ batchScheduler:
 * Pass the `--set batchScheduler.name=volcano` flag when running on the command line:
 ```shell
 # Install the Helm chart with the --batch-scheduler=volcano flag
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2 --set batchScheduler.name=volcano
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0 --set batchScheduler.name=volcano
 ```
 
 ### Step 4: Install a RayCluster with the Volcano scheduler
@@ -45,7 +45,7 @@ The RayCluster custom resource must include the `ray.io/scheduler-name: volcano`
 ```shell
 # Path: kuberay/ray-operator/config/samples
 # Includes label `ray.io/scheduler-name: volcano` in the metadata.labels
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-cluster.volcano-scheduler.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-cluster.volcano-scheduler.yaml
 kubectl apply -f ray-cluster.volcano-scheduler.yaml
 
 # Check the RayCluster
@@ -113,7 +113,7 @@ Next, create a RayCluster with a head node (1 CPU + 2Gi of RAM) and two workers
 ```shell
 # Path: kuberay/ray-operator/config/samples
 # Includes  the `ray.io/scheduler-name: volcano` and `volcano.sh/queue-name: kuberay-test-queue` labels in the metadata.labels
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-cluster.volcano-scheduler-queue.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-cluster.volcano-scheduler-queue.yaml
 kubectl apply -f ray-cluster.volcano-scheduler-queue.yaml
 ```
 
@@ -332,7 +332,7 @@ Starting with KubeRay 1.5.0, KubeRay supports gang scheduling for RayJob custom
 First, create a queue with a capacity of 4 CPUs and 6Gi of RAM and RayJob a with a head node (1 CPU + 2Gi of RAM), two workers (1 CPU + 1Gi of RAM each) and a submitter pod (0.5 CPU + 200Mi of RAM), for a total of 3500m CPU and 4296Mi of RAM
 
 ```shell
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/refs/tags/v1.5.0-rc.0/ray-operator/config/samples/ray-job.volcano-scheduler-queue.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-job.volcano-scheduler-queue.yaml
 kubectl apply -f ray-job.volcano-scheduler-queue.yaml
 ```
 

@@ -29,7 +29,7 @@ See [Get Started](https://yunikorn.apache.org/docs/) for Apache YuniKorn install
 When installing KubeRay operator using Helm, pass the `--set batchScheduler.name=yunikorn` flag at the command line:
 
 ```shell
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2 --set batchScheduler.name=yunikorn
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0 --set batchScheduler.name=yunikorn
 ```
 
 ## Step 4: Use Apache YuniKorn for gang scheduling

@@ -64,7 +64,7 @@ Follow [this document](kuberay-operator-deploy) to install the latest stable Kub
 ### Step 3: Create a RayCluster custom resource with autoscaling enabled
 
 ```bash
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-cluster.autoscaler.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-cluster.autoscaler.yaml
 ```
 
 ### Step 4: Verify the Kubernetes cluster status
@@ -87,7 +87,7 @@ kubectl get configmaps
 ```
 
 The RayCluster has one head Pod and zero worker Pods. The head Pod has two containers: a Ray head container and a Ray Autoscaler sidecar container.
-Additionally, the [ray-cluster.autoscaler.yaml](https://github.com/ray-project/kuberay/blob/v1.4.2/ray-operator/config/samples/ray-cluster.autoscaler.yaml) includes a ConfigMap named `ray-example` that contains two Python scripts: `detached_actor.py` and `terminate_detached_actor.py`.
+Additionally, the [ray-cluster.autoscaler.yaml](https://github.com/ray-project/kuberay/blob/v1.5.0/ray-operator/config/samples/ray-cluster.autoscaler.yaml) includes a ConfigMap named `ray-example` that contains two Python scripts: `detached_actor.py` and `terminate_detached_actor.py`.
 
 * `detached_actor.py` is a Python script that creates a detached actor which requires 1 CPU.
   ```py
@@ -247,7 +247,7 @@ kubectl logs $HEAD_POD -c autoscaler | tail -n 20
 
 ```bash
 # Delete RayCluster and ConfigMap
-kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-cluster.autoscaler.yaml
+kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-cluster.autoscaler.yaml
 
 # Uninstall the KubeRay operator
 helm uninstall kuberay-operator
@@ -259,7 +259,7 @@ kind delete cluster
 (kuberay-autoscaling-config)=
 ## KubeRay Autoscaling Configurations
 
-The [ray-cluster.autoscaler.yaml](https://github.com/ray-project/kuberay/blob/v1.4.2/ray-operator/config/samples/ray-cluster.autoscaler.yaml) used in the quickstart example contains detailed comments about the configuration options.
+The [ray-cluster.autoscaler.yaml](https://github.com/ray-project/kuberay/blob/v1.5.0/ray-operator/config/samples/ray-cluster.autoscaler.yaml) used in the quickstart example contains detailed comments about the configuration options.
 ***It's recommended to read this section in conjunction with the YAML file.***
 
 ### 1. Enabling autoscaling
@@ -333,7 +333,7 @@ If you omit `rayStartParams` and want to use autoscaling, the autoscaling image
 The Ray Autoscaler reads the `rayStartParams` field or the Ray container's resource limits in the RayCluster custom resource specification to determine the Ray Pod's resource requirements.
 The information regarding the number of CPUs is essential for the Ray Autoscaler to scale the cluster.
 Therefore, without this information, the Ray Autoscaler reports an error and fails to start.
-Take [ray-cluster.autoscaler.yaml](https://github.com/ray-project/kuberay/blob/v1.4.2/ray-operator/config/samples/ray-cluster.autoscaler.yaml) as an example below:
+Take [ray-cluster.autoscaler.yaml](https://github.com/ray-project/kuberay/blob/v1.5.0/ray-operator/config/samples/ray-cluster.autoscaler.yaml) as an example below:
 
 * If users set `num-cpus` in `rayStartParams`, Ray Autoscaler would work regardless of the resource limits on the container.
 * If users don't set `rayStartParams`, the Ray container must have a specified CPU resource limit.
@@ -458,7 +458,7 @@ Node: 2d5fd3d4337ba5b5a8c3106c572492abb9a8de2dee9da7f6c24c1346
 2. **Stability**
 Autoscaler V2 makes significant improvements to idle node handling. The V1 autoscaler could stop nodes that became active during termination processing, potentially failing tasks or actors. V2 uses Ray's graceful draining mechanism, which safely stops idle nodes without disrupting ongoing work.
 
-[ray-cluster.autoscaler-v2.yaml](https://github.com/ray-project/kuberay/blob/v1.4.2/ray-operator/config/samples/ray-cluster.autoscaler-v2.yaml) is an example YAML file of a RayCluster with Autoscaler V2 enabled that works with the latest KubeRay version.
+[ray-cluster.autoscaler-v2.yaml](https://github.com/ray-project/kuberay/blob/v1.5.0/ray-operator/config/samples/ray-cluster.autoscaler-v2.yaml) is an example YAML file of a RayCluster with Autoscaler V2 enabled that works with the latest KubeRay version.
 
 If you're using KubeRay >= 1.4.0, enable V2 by setting `RayCluster.spec.autoscalerOptions.version: v2`.
 

@@ -83,8 +83,8 @@ In a cluster without the Ray Operator Addon enabled, KubeRay can be manually ins
 ```sh
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 
-# Install both CRDs and KubeRay operator v1.4.2.
-helm install kuberay-operator kuberay/kuberay-operator --version 1.4.2
+# Install both CRDs and KubeRay operator v1.5.0.
+helm install kuberay-operator kuberay/kuberay-operator --version 1.5.0
 ```
 
 GKE provides a [validating and mutating webhook](https://github.com/ai-on-gke/kuberay-tpu-webhook) to handle TPU Pod scheduling and bootstrap certain environment variables used for [JAX](https://github.com/google/jax) initialization. The Ray TPU webhook requires a KubeRay operator version of at least v1.1.0. GKE automatically installs the Ray TPU webhook through the [Ray Operator Addon](https://cloud.google.com/kubernetes-engine/docs/add-on/ray-on-gke/how-to/enable-ray-on-gke) with GKE versions 1.30.0-gke.1747000 or later.

@@ -72,7 +72,7 @@ gsutil iam ch serviceAccount:my-iam-sa@my-project-id.iam.gserviceaccount.com:rol
 You can download the RayCluster YAML manifest for this tutorial with `curl` as follows:
 
 ```bash
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.4.2/ray-operator/config/samples/ray-cluster.gke-bucket.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.5.0/ray-operator/config/samples/ray-cluster.gke-bucket.yaml
 ```
 
 The key parts are the following lines:
-Original file line number
+Diff line change
@@ Expand Up @@
     ```sh
     # Deploy a sample RayCluster CR from the KubeRay Helm chart repo:
-    helm install raycluster kuberay/ray-cluster --version 1.4.2
+    helm install raycluster kuberay/ray-cluster --version 1.5.0
     ```
@@ Expand Down @@