Remote cluster config on "deployments" tab, update image versions

scripts/envs/audio-to-text

@@ -12,8 +12,7 @@ export APP_RUN_COMMAND="streamlit run whisper_client.py"
 export INIT_CONTAINER="quay.io/redhat-ai-dev/whisper-small:latest"
 
 # https://github.com/redhat-ai-dev/developer-images/blob/main/model-servers/whispercpp/1.7.6/Containerfile
-# https://github.com/ggerganov/whisper.cpp/blob/v1.7.6/examples/server/README.md
-export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/whispercpp:1.7.6"
+export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/whispercpp:1.8.0"
 export MODEL_SERVICE_DESC="Simple HTTP server where WAV Files are passed to the inference model via HTTP requests" 
 export MODEL_SERVICE_SRC="https://github.com/containers/ai-lab-recipes/tree/main/model_servers/whispercpp"
 

scripts/envs/chatbot

@@ -11,15 +11,15 @@ export APP_RUN_COMMAND="streamlit run chatbot_ui.py"
 # https://github.com/redhat-ai-dev/developer-images/tree/main/models/granite-3.1-8b-instruct-gguf
 export INIT_CONTAINER="quay.io/redhat-ai-dev/granite-3.1-8b-instruct-gguf:latest"
 
-# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8
-export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/llamacpp_python:0.3.8"
+# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16
+export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/llamacpp_python:0.3.16"
 export MODEL_SERVICE_DESC="A Python binding of LLM inference in C/C++ with minimal setup"
-export MODEL_SERVICE_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8"
+export MODEL_SERVICE_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16"
 
 # https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4
-export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.8.4"
+export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.11.0"
 export VLLM_DESC="A high throughput, memory efficient inference and serving engine with GPU support for LLMs in OpenShift"
-export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4"
+export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0"
 
 # https://huggingface.co/ibm-granite/granite-3.1-8b-instruct
 export LLM_MODEL_NAME="ibm-granite/granite-3.1-8b-instruct"

scripts/envs/codegen

@@ -11,15 +11,15 @@ export APP_RUN_COMMAND="streamlit run codegen-app.py"
 # https://github.com/redhat-ai-dev/developer-images/tree/main/models/mistral-7b-instruct-v0.2
 export INIT_CONTAINER="quay.io/redhat-ai-dev/mistral-7b-instruct-v0.2:latest"
 
-# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8
-export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/llamacpp_python:0.3.8"
+# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16
+export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/llamacpp_python:0.3.16"
 export MODEL_SERVICE_DESC="A Python binding of LLM inference in C/C++ with minimal setup"
-export MODEL_SERVICE_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8"
+export MODEL_SERVICE_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16"
 
-# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4
-export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.8.4"
+# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0
+export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.11.0"
 export VLLM_DESC="A high throughput, memory efficient inference and serving engine with GPU support for LLMs in OpenShift"
-export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4"
+export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0"
 
 # https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-AWQ
 export LLM_MODEL_NAME="TheBloke/Mistral-7B-Instruct-v0.2-AWQ"

scripts/envs/model-server

@@ -7,10 +7,10 @@ export APP_DESC="Deploy a granite-3.1 8b model with a vLLM server. While no appl
 export APP_SUMMARY="A granite-3.1 8b model server deployment."
 export APP_TAGS='["ai", "vllm", "modelserver"]'
 
-# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4
-export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.8.4"
+# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0
+export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.11.0"
 export VLLM_DESC="A high throughput, memory efficient inference and serving engine with GPU support for LLMs in OpenShift"
-export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4"
+export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0"
 
 # https://huggingface.co/ibm-granite/granite-3.1-8b-instruct
 export LLM_MODEL_NAME="ibm-granite/granite-3.1-8b-instruct"

scripts/envs/rag

@@ -11,15 +11,15 @@ export APP_RUN_COMMAND="streamlit run rag_app.py"
 # https://github.com/redhat-ai-dev/developer-images/tree/main/models/granite-3.1-8b-instruct-gguf
 export INIT_CONTAINER="quay.io/redhat-ai-dev/granite-3.1-8b-instruct-gguf:latest"
 
-# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8
-export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/llamacpp_python:0.3.8"
+# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16
+export MODEL_SERVICE_CONTAINER="quay.io/redhat-ai-dev/llamacpp_python:0.3.16"
 export MODEL_SERVICE_DESC="A Python binding of LLM inference in C/C++ with minimal setup"
-export MODEL_SERVICE_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8"
+export MODEL_SERVICE_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16"
 
-# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4
-export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.8.4"
+# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0
+export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.11.0"
 export VLLM_DESC="A high throughput, memory efficient inference and serving engine with GPU support for LLMs in OpenShift"
-export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4"
+export VLLM_SRC="https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0"
 
 # https://huggingface.co/ibm-granite/granite-3.1-8b-instruct
 export LLM_MODEL_NAME="ibm-granite/granite-3.1-8b-instruct"

scripts/util

@@ -126,4 +126,4 @@ function apply-configurations() {
     source $ROOT_DIR/properties
     cat $DEST/template.yaml | envsubst >$DEST/new-template.yaml
     mv $DEST/new-template.yaml $DEST/template.yaml
-}
+}

skeleton/template.yaml

@@ -451,6 +451,15 @@ spec:
         - imageName
         # SED_APP_SUPPORT_END
         - namespace
+      ui:order:
+        - imageRegistry
+        - imageOrg
+        - imageName
+        - namespace
+        - deployArgoCDApplicationOnRemoteCluster
+        - remoteClusterAPIUrl
+        - remoteClusterDeploymentNamespace
+        - rhoaiSelected
       properties:
         # SED_APP_SUPPORT_START
         imageRegistry:
@@ -477,6 +486,11 @@ spec:
           ui:autofocus: true
           ui:options:
             rows: 5
+        deployArgoCDApplicationOnRemoteCluster:
+          title: Deploy ArgoCD Application on Remote Cluster?
+          type: boolean
+          default: false
+          ui:help: "Check this to deploy the ArgoCD application to a remote cluster"
         # SED_APP_SUPPORT_START
         rhoaiSelected:
           title: Create Workbench for OpenShift AI
@@ -485,6 +499,26 @@ spec:
           default: false
           ui:help: If you select this field, you must ensure that Red Hat OpenShift AI has been installed on your cluster.
           # SED_APP_SUPPORT_END
+      dependencies:
+        deployArgoCDApplicationOnRemoteCluster:
+          oneOf:
+            - required:
+                - remoteClusterAPIUrl
+                - remoteClusterDeploymentNamespace
+              properties:
+                deployArgoCDApplicationOnRemoteCluster:
+                  const: true
+                remoteClusterAPIUrl:
+                  title: Remote Cluster API URL
+                  type: string
+                  ui:help: "Kube API URL of remote cluster"
+                remoteClusterDeploymentNamespace:
+                  title: Remote Cluster Deployment Namespace
+                  type: string
+                  ui:help: "The namespace of the remote cluster that the argoCD application will be deployed"
+            - properties:
+                deployArgoCDApplicationOnRemoteCluster:
+                  const: false
   # These steps are executed in the scaffolder backend, using data that we gathered
   # via the parameters above.
   steps:

templates/audio-to-text/template.yaml

@@ -254,6 +254,15 @@ spec:
         - imageName
         # SED_APP_SUPPORT_END
         - namespace
+      ui:order:
+        - imageRegistry
+        - imageOrg
+        - imageName
+        - namespace
+        - deployArgoCDApplicationOnRemoteCluster
+        - remoteClusterAPIUrl
+        - remoteClusterDeploymentNamespace
+        - rhoaiSelected
       properties:
         # SED_APP_SUPPORT_START
         imageRegistry:
@@ -280,6 +289,11 @@ spec:
           ui:autofocus: true
           ui:options:
             rows: 5
+        deployArgoCDApplicationOnRemoteCluster:
+          title: Deploy ArgoCD Application on Remote Cluster?
+          type: boolean
+          default: false
+          ui:help: "Check this to deploy the ArgoCD application to a remote cluster"
         # SED_APP_SUPPORT_START
         rhoaiSelected:
           title: Create Workbench for OpenShift AI
@@ -288,6 +302,26 @@ spec:
           default: false
           ui:help: If you select this field, you must ensure that Red Hat OpenShift AI has been installed on your cluster.
           # SED_APP_SUPPORT_END
+      dependencies:
+        deployArgoCDApplicationOnRemoteCluster:
+          oneOf:
+            - required:
+                - remoteClusterAPIUrl
+                - remoteClusterDeploymentNamespace
+              properties:
+                deployArgoCDApplicationOnRemoteCluster:
+                  const: true
+                remoteClusterAPIUrl:
+                  title: Remote Cluster API URL
+                  type: string
+                  ui:help: "Kube API URL of remote cluster"
+                remoteClusterDeploymentNamespace:
+                  title: Remote Cluster Deployment Namespace
+                  type: string
+                  ui:help: "The namespace of the remote cluster that the argoCD application will be deployed"
+            - properties:
+                deployArgoCDApplicationOnRemoteCluster:
+                  const: false
   # These steps are executed in the scaffolder backend, using data that we gathered
   # via the parameters above.
   steps:
@@ -318,7 +352,7 @@ spec:
           appContainer: ${{ 'quay.io/redhat-ai-dev/ai-template-bootstrap-app:latest' if parameters.hostType === 'GitHub' else 'quay.io/redhat-ai-dev/audio-to-text:latest' }}
           appPort: 8501
           appRunCommand: "streamlit run whisper_client.py"
-          modelServiceContainer: quay.io/redhat-ai-dev/whispercpp:1.7.6
+          modelServiceContainer: quay.io/redhat-ai-dev/whispercpp:1.8.0
           modelServicePort: 8001
           customModelName: ${{ steps['fetch-model-from-catalog'].output.entity.metadata.name if parameters.modelServer === 'choose-from-the-catalog' else parameters.modelName }}
           modelName: ggerganov/whisper.cpp
@@ -413,7 +447,7 @@ spec:
           modelPath: "/model/model.file"
           appContainer: ${{ 'quay.io/redhat-ai-dev/ai-template-bootstrap-app:latest' if parameters.hostType === 'GitHub' else 'quay.io/redhat-ai-dev/audio-to-text:latest' }}
           appPort: 8501
-          modelServiceContainer: quay.io/redhat-ai-dev/whispercpp:1.7.6
+          modelServiceContainer: quay.io/redhat-ai-dev/whispercpp:1.8.0
           modelServicePort: 8001
           existingModelServer: ${{ true if parameters.modelServer === 'Bring you own model server' else (true if parameters.modelServer === 'choose-from-the-catalog' else false) }}
           # SED_EXISTING_START

templates/chatbot/template.yaml

@@ -206,7 +206,7 @@ spec:
                   const: vLLM
                 modelServerDescription:
                   type: 'null'
-                  description: A high throughput, memory efficient inference and serving engine with GPU support for LLMs in OpenShift. If you choose vLLM, ensure that your cluster has Nvidia GPU supported (with compute capability 7.0 or higher). Also, it should have enough CPU & memory resources for the model you would like to work with. | [Learn more](https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4)
+                  description: A high throughput, memory efficient inference and serving engine with GPU support for LLMs in OpenShift. If you choose vLLM, ensure that your cluster has Nvidia GPU supported (with compute capability 7.0 or higher). Also, it should have enough CPU & memory resources for the model you would like to work with. | [Learn more](https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0)
                 modelNameDeployed:
                   title: Model Name
                   description: Text Generation | Apache-2.0 | [Learn more](https://huggingface.co/ibm-granite/granite-3.1-8b-instruct)
@@ -220,7 +220,7 @@ spec:
                   const: llama.cpp
                 modelServerDescription:
                   type: 'null'
-                  description: A Python binding of LLM inference in C/C++ with minimal setup. | [Learn more](https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8)
+                  description: A Python binding of LLM inference in C/C++ with minimal setup. | [Learn more](https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16)
                 modelNameDeployed:
                   title: Model Name
                   description: Text Generation | Apache-2.0 | [Learn more](https://huggingface.co/ibm-granite/granite-3.1-8b-instruct)
@@ -286,6 +286,15 @@ spec:
         - imageName
         # SED_APP_SUPPORT_END
         - namespace
+      ui:order:
+        - imageRegistry
+        - imageOrg
+        - imageName
+        - namespace
+        - deployArgoCDApplicationOnRemoteCluster
+        - remoteClusterAPIUrl
+        - remoteClusterDeploymentNamespace
+        - rhoaiSelected
       properties:
         # SED_APP_SUPPORT_START
         imageRegistry:
@@ -312,6 +321,11 @@ spec:
           ui:autofocus: true
           ui:options:
             rows: 5
+        deployArgoCDApplicationOnRemoteCluster:
+          title: Deploy ArgoCD Application on Remote Cluster?
+          type: boolean
+          default: false
+          ui:help: "Check this to deploy the ArgoCD application to a remote cluster"
         # SED_APP_SUPPORT_START
         rhoaiSelected:
           title: Create Workbench for OpenShift AI
@@ -320,6 +334,26 @@ spec:
           default: false
           ui:help: If you select this field, you must ensure that Red Hat OpenShift AI has been installed on your cluster.
           # SED_APP_SUPPORT_END
+      dependencies:
+        deployArgoCDApplicationOnRemoteCluster:
+          oneOf:
+            - required:
+                - remoteClusterAPIUrl
+                - remoteClusterDeploymentNamespace
+              properties:
+                deployArgoCDApplicationOnRemoteCluster:
+                  const: true
+                remoteClusterAPIUrl:
+                  title: Remote Cluster API URL
+                  type: string
+                  ui:help: "Kube API URL of remote cluster"
+                remoteClusterDeploymentNamespace:
+                  title: Remote Cluster Deployment Namespace
+                  type: string
+                  ui:help: "The namespace of the remote cluster that the argoCD application will be deployed"
+            - properties:
+                deployArgoCDApplicationOnRemoteCluster:
+                  const: false
   # These steps are executed in the scaffolder backend, using data that we gathered
   # via the parameters above.
   steps:
@@ -350,15 +384,15 @@ spec:
           appContainer: ${{ 'quay.io/redhat-ai-dev/ai-template-bootstrap-app:latest' if parameters.hostType === 'GitHub' else 'quay.io/redhat-ai-dev/chatbot:latest' }}
           appPort: 8501
           appRunCommand: "streamlit run chatbot_ui.py"
-          modelServiceContainer: quay.io/redhat-ai-dev/llamacpp_python:0.3.8
+          modelServiceContainer: quay.io/redhat-ai-dev/llamacpp_python:0.3.16
           modelServicePort: 8001
           customModelName: ${{ steps['fetch-model-from-catalog'].output.entity.metadata.name if parameters.modelServer === 'choose-from-the-catalog' else parameters.modelName }}
           modelName: ibm-granite/granite-3.1-8b-instruct
           modelSrc: https://huggingface.co/ibm-granite/granite-3.1-8b-instruct
           modelServerName: ${{ parameters.modelServer }}
           customModelAndModelServerSelected: ${{ true if parameters.modelServer === 'Bring you own model server' else (true if parameters.modelServer === 'choose-from-the-catalog' else false) }}
-          modelServiceSrcVLLM: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4
-          modelServiceSrcOther: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8
+          modelServiceSrcVLLM: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0
+          modelServiceSrcOther: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16
     # Renders all the template variables into the files and directory names and content, and places the result in the workspace.
     - id: fetch-skeleton
       name: Fetch Skeleton
@@ -445,12 +479,12 @@ spec:
           modelPath: "/model/model.file"
           appContainer: ${{ 'quay.io/redhat-ai-dev/ai-template-bootstrap-app:latest' if parameters.hostType === 'GitHub' else 'quay.io/redhat-ai-dev/chatbot:latest' }}
           appPort: 8501
-          modelServiceContainer: quay.io/redhat-ai-dev/llamacpp_python:0.3.8
+          modelServiceContainer: quay.io/redhat-ai-dev/llamacpp_python:0.3.16
           modelServicePort: 8001
           # SED_LLM_SERVER_START
           # for vllm
           vllmSelected: ${{ parameters.modelServer === 'vLLM' }}
-          vllmModelServiceContainer: quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.8.4
+          vllmModelServiceContainer: quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.11.0
           modelName: ${{ parameters.modelName if parameters.modelServer === 'Bring you own model server' else (steps['fetch-model-from-catalog'].output.entity.metadata.name if parameters.modelServer === 'choose-from-the-catalog' else 'ibm-granite/granite-3.1-8b-instruct') }}
           modelSrc: https://huggingface.co/ibm-granite/granite-3.1-8b-instruct
           maxModelLength: 4096
@@ -470,8 +504,8 @@ spec:
           dbRequired: false
           supportApp: true
           modelServerName: ${{ parameters.modelServer }}
-          modelServiceSrcVLLM: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.8.4
-          modelServiceSrcOther: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.8
+          modelServiceSrcVLLM: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.11.0
+          modelServiceSrcOther: https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/llamacpp_python/0.3.16
           imageRegistry: ${{ parameters.imageRegistry }}
           imageOrg: ${{ parameters.imageOrg }}
           imageName: ${{ parameters.imageName }}