Upgrade Granite Model Template to granite-3.1

scripts/envs/model-server

@@ -3,21 +3,21 @@
 # env var used only for this sample you can list it here
 export APP_NAME="model-server"
 export APP_DISPLAY_NAME="Model Server, No Application"
-export APP_DESC="Deploy a granite-3.0 8b model with a vLLM server. While no application is configured, this model server can be utilized in other Software Templates, like a Chatbot Application for instance."
-export APP_SUMMARY="A granite-3.0 8b model server deployment."
+export APP_DESC="Deploy a granite-3.1 8b model with a vLLM server. While no application is configured, this model server can be utilized in other Software Templates, like a Chatbot Application for instance."
+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/0.6.4
-export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.6.4"
+# https://github.com/redhat-ai-dev/developer-images/tree/main/model-servers/vllm/0.6.6
+export VLLM_CONTAINER="quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.6.6"
 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/rh-aiservices-bu/llm-on-openshift/tree/main/llm-servers/vllm/gpu"
 
-# https://huggingface.co/ibm-granite/granite-3.0-8b-instruct
-export LLM_MODEL_NAME="ibm-granite/granite-3.0-8b-instruct"
+# https://huggingface.co/ibm-granite/granite-3.1-8b-instruct
+export LLM_MODEL_NAME="ibm-granite/granite-3.1-8b-instruct"
 export LLM_MAX_MODEL_LEN=4096
 export LLM_MODEL_CLASSIFICATION="Text Generation"
 export LLM_MODEL_LICENSE="Apache-2.0"
-export LLM_MODEL_SRC="https://huggingface.co/ibm-granite/granite-3.0-8b-instruct"
+export LLM_MODEL_SRC="https://huggingface.co/ibm-granite/granite-3.1-8b-instruct"
 
 #model configuration
 export SUPPORT_LLM=true

skeleton/source-repo/.tekton/docker-pull-request.yaml

@@ -6,13 +6,12 @@ metadata:
     pipelinesascode.tekton.dev/on-event: "[pull_request]"
     pipelinesascode.tekton.dev/on-target-branch: "[${{ values.defaultBranch }}]"
     pipelinesascode.tekton.dev/max-keep-runs: "2"
-    pipelinesascode.tekton.dev/pipeline: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/pipelines/docker-build-ai-rhdh.yaml"
+    pipelinesascode.tekton.dev/pipeline: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/pipelines/docker-build-ai-rhdh-pull-request.yaml"
     pipelinesascode.tekton.dev/task-0: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/init.yaml"
     pipelinesascode.tekton.dev/task-1: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/git-clone.yaml"
     pipelinesascode.tekton.dev/task-2: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/buildah-ai-rhdh.yaml"
-    pipelinesascode.tekton.dev/task-3: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/update-deployment.yaml"
-    pipelinesascode.tekton.dev/task-4: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/show-sbom-rhdh.yaml"
-    pipelinesascode.tekton.dev/task-5: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/summary.yaml"
+    pipelinesascode.tekton.dev/task-3: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/show-sbom-rhdh.yaml"
+    pipelinesascode.tekton.dev/task-4: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/summary.yaml"
   labels:
     argocd/app-name: ${{ values.name }}
     janus-idp.io/tekton: ${{ values.name }}
@@ -33,12 +32,8 @@ spec:
       value: ${{ values.buildContext }}
     - name: revision
       value: '{{revision}}'
-    - name: event-type
-      value: '{{event_type}}'
-    - name: gitops-auth-secret-name
-      value: ${{ values.gitopsSecretName }}
   pipelineRef:
-    name: docker-build-ai-rhdh
+    name: docker-build-ai-rhdh-pull-request
   workspaces:
     - name: git-auth
       secret:

skeleton/source-repo/.tekton/docker-push.yaml

@@ -6,11 +6,11 @@ metadata:
     pipelinesascode.tekton.dev/on-event: "[push]"
     pipelinesascode.tekton.dev/on-target-branch: "[${{ values.defaultBranch }}]"
     pipelinesascode.tekton.dev/max-keep-runs: "2"
-    pipelinesascode.tekton.dev/pipeline: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/pipelines/docker-build-ai-rhdh.yaml"
+    pipelinesascode.tekton.dev/pipeline: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/pipelines/docker-build-ai-rhdh-push-gitops.yaml"
     pipelinesascode.tekton.dev/task-0: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/init.yaml"
     pipelinesascode.tekton.dev/task-1: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/git-clone.yaml"
     pipelinesascode.tekton.dev/task-2: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/buildah-ai-rhdh.yaml"
-    pipelinesascode.tekton.dev/task-3: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/update-deployment.yaml"
+    pipelinesascode.tekton.dev/task-3: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/update-deployment-gitops.yaml"
     pipelinesascode.tekton.dev/task-4: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/show-sbom-rhdh.yaml"
     pipelinesascode.tekton.dev/task-5: "https://raw.githubusercontent.com/redhat-ai-dev/rhdh-pipelines/main/pac/tasks/summary.yaml"
   labels:
@@ -33,16 +33,17 @@ spec:
       value: ${{ values.buildContext }}
     - name: revision
       value: '{{revision}}'
-    - name: event-type
-      value: '{{event_type}}'
     - name: gitops-auth-secret-name
       value: ${{ values.gitopsSecretName }}
   pipelineRef:
-    name: docker-build-ai-rhdh
+    name: docker-build-ai-rhdh-push-gitops
   workspaces:
     - name: git-auth
       secret:
         secretName: "{{ git_auth_secret }}"
+    - name: gitops-auth
+      secret:
+        secretName: $(params.gitops-auth-secret-name)
     - name: workspace
       volumeClaimTemplate:
         spec:

templates/model-server/docs/index.md

@@ -23,4 +23,4 @@ As a user you will be able to customize each section of the template, such as:
 - Application deployment namespace
 
 
-For more information please check out our general template documentation in our [source code repository](https://github.com/redhat-ai-dev/ai-lab-template). If you are interested in the ibm-granite/granite-3.0-8b-instruct model being used for this template, refer to the hugging face [page](https://huggingface.co/ibm-granite/granite-3.0-8b-instruct)!
+For more information please check out our general template documentation in our [source code repository](https://github.com/redhat-ai-dev/ai-lab-template). If you are interested in the ibm-granite/granite-3.1-8b-instruct model being used for this template, refer to the hugging face [page](https://huggingface.co/ibm-granite/granite-3.1-8b-instruct)!

templates/model-server/template.yaml

@@ -4,7 +4,7 @@ kind: Template
 metadata:
   name: model-server
   title: Model Server, No Application
-  description: Deploy a granite-3.0 8b model with a vLLM server. While no application is configured, this model server can be utilized in other Software Templates, like a Chatbot Application for instance.
+  description: Deploy a granite-3.1 8b model with a vLLM server. While no application is configured, this model server can be utilized in other Software Templates, like a Chatbot Application for instance.
   tags: ["ai", "vllm", "modelserver"]
   annotations:
     backstage.io/techdocs-ref: dir:.
@@ -58,11 +58,11 @@ spec:
                   const: vLLM
                 modelNameDeployed:
                   title: Model Name
-                  description: Text Generation | Apache-2.0 | [Learn more](https://huggingface.co/ibm-granite/granite-3.0-8b-instruct)
-                  default: ibm-granite/granite-3.0-8b-instruct
+                  description: Text Generation | Apache-2.0 | [Learn more](https://huggingface.co/ibm-granite/granite-3.1-8b-instruct)
+                  default: ibm-granite/granite-3.1-8b-instruct
                   type: string
                   enum:
-                    - ibm-granite/granite-3.0-8b-instruct
+                    - ibm-granite/granite-3.1-8b-instruct
             # SED_LLM_SERVER_END
 
     - title: Application Repository Information
@@ -163,8 +163,8 @@ spec:
           # SED_LLM_SERVER_START
           # for vllm
           vllmSelected: ${{ parameters.modelServer === 'vLLM' }}
-          vllmModelServiceContainer: quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.6.4
-          modelName: ${{ parameters.modelName if parameters.modelServer === 'Existing model server' else 'ibm-granite/granite-3.0-8b-instruct' }}
+          vllmModelServiceContainer: quay.io/redhat-ai-dev/vllm-openai-ubi9:v0.6.6
+          modelName: ${{ parameters.modelName if parameters.modelServer === 'Existing model server' else 'ibm-granite/granite-3.1-8b-instruct' }}
           maxModelLength: 4096
           # SED_LLM_SERVER_END
           existingModelServer: ${{ parameters.modelServer === 'Existing model server' }}