Merge pull request #1714 from slp/install-virglrenderer

container-images: add virglrenderer to vulkan

.gitignore

@@ -19,3 +19,5 @@ __pycache__/
 coverage.*
 htmlcov/
 .idea/
+.hypothesis/
+uv.lock

.packit-copr-rpm.sh

@@ -9,8 +9,7 @@ set -exo pipefail
 # Extract version from pyproject.toml instead of setup.py
 VERSION=$(awk -F'[""]' ' /^\s*version\s*/ {print $(NF-1)}' pyproject.toml )
 
-#SPEC_FILE=rpm-next/ramalama.spec
-SPEC_FILE=rpm/python-ramalama.spec
+SPEC_FILE=rpm/ramalama.spec
 
 # RPM Spec modifications
 

.packit.yaml

@@ -7,21 +7,11 @@ upstream_tag_template: v{version}
 packages:
   ramalama-fedora:
     pkg_tool: fedpkg
-    downstream_package_name: python-ramalama
-    specfile_path: rpm/python-ramalama.spec
+    downstream_package_name: ramalama
+    specfile_path: rpm/ramalama.spec
   ramalama-centos:
-    downstream_package_name: python-ramalama
-    specfile_path: rpm/python-ramalama.spec
-
-# Uncomment when we get this approved.
-# packages:
-#   ramalama-fedora:
-#     pkg_tool: fedpkg
-#     downstream_package_name: ramalama
-#     specfile_path: rpm-next/ramalama.spec
-#   ramalama-centos:
-#     downstream_package_name: ramalama
-#     specfile_path: rpm-next/ramalama.spec
+    downstream_package_name: ramalama
+    specfile_path: rpm/ramalama.spec
 
 srpm_build_deps:
   - make
@@ -89,7 +79,7 @@ jobs:
     dist_git_branches: &fedora_targets
       - fedora-all
       - epel10
-      - epel9
+      - epel10.0
 
   - job: koji_build
     trigger: commit
@@ -102,4 +92,4 @@ jobs:
     dist_git_branches:
       - fedora-branched # rawhide updates are created automatically
       - epel10
-      - epel9
+      - epel10.0

.tekton/asahi-llama-server/asahi-llama-server-pull-request.yaml

@@ -0,0 +1,48 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-llama-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi-llama-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/asahi:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/asahi-llama-server/asahi-llama-server-push.yaml

@@ -0,0 +1,45 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-llama-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi-llama-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/asahi:{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/asahi-rag/asahi-rag-pull-request.yaml

@@ -0,0 +1,48 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-rag-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi-rag:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/asahi:on-pr-{{revision}}
+    - GPU=cpu
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/asahi-rag/asahi-rag-push.yaml

@@ -0,0 +1,45 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-rag-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi-rag:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/asahi:{{revision}}
+    - GPU=cpu
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/asahi-whisper-server/asahi-whisper-server-pull-request.yaml

@@ -0,0 +1,48 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-whisper-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi-whisper-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/asahi:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/asahi-whisper-server/asahi-whisper-server-push.yaml

@@ -0,0 +1,45 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-whisper-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi-whisper-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/asahi:{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/asahi/asahi-pull-request.yaml

@@ -0,0 +1,42 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/asahi/Containerfile
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/asahi/asahi-push.yaml

@@ -0,0 +1,39 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: asahi
+    pipelines.appstudio.openshift.io/type: build
+  name: asahi-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/asahi:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/asahi/Containerfile
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/bats/bats-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: bats
@@ -28,8 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
-    - linux/arm64
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
   - name: dockerfile
     value: container-images/bats/Containerfile
   pipelineRef:

.tekton/bats/bats-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: bats
@@ -25,8 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/bats:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
-    - linux/arm64
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
   - name: dockerfile
     value: container-images/bats/Containerfile
   pipelineRef:

.tekton/cann-llama-server/cann-llama-server-pull-request.yaml

@@ -0,0 +1,48 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-llama-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann-llama-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/cann:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cann-llama-server/cann-llama-server-push.yaml

@@ -0,0 +1,45 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-llama-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann-llama-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/cann:{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cann-rag/cann-rag-pull-request.yaml

@@ -0,0 +1,48 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-rag-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann-rag:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/cann:on-pr-{{revision}}
+    - GPU=cpu
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cann-rag/cann-rag-push.yaml

@@ -0,0 +1,45 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-rag-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann-rag:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/cann:{{revision}}
+    - GPU=cpu
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cann-whisper-server/cann-whisper-server-pull-request.yaml

@@ -0,0 +1,48 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-whisper-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann-whisper-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/cann:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cann-whisper-server/cann-whisper-server-push.yaml

@@ -0,0 +1,45 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-whisper-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann-whisper-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/cann:{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cann/cann-pull-request.yaml

@@ -0,0 +1,42 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/cann/Containerfile
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cann/cann-push.yaml

@@ -0,0 +1,39 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: cann
+    pipelines.appstudio.openshift.io/type: build
+  name: cann-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/cann:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/cann/Containerfile
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/cuda-llama-server/cuda-llama-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda-llama-server
@@ -28,7 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/cuda-llama-server/cuda-llama-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda-llama-server
@@ -25,7 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/cuda-llama-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/cuda-rag/cuda-rag-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda-rag
@@ -28,7 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/cuda-rag/cuda-rag-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda-rag
@@ -25,7 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/cuda-rag:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/cuda-whisper-server/cuda-whisper-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda-whisper-server
@@ -28,7 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/cuda-whisper-server/cuda-whisper-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda-whisper-server
@@ -25,7 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/cuda-whisper-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/cuda/cuda-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda
@@ -28,7 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
   - name: dockerfile
     value: container-images/cuda/Containerfile
   pipelineRef:

.tekton/cuda/cuda-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: cuda
@@ -25,7 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/cuda:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
   - name: dockerfile
     value: container-images/cuda/Containerfile
   pipelineRef:

.tekton/integration/pipelines/bats-integration.yaml

@@ -0,0 +1,66 @@
+kind: Pipeline
+apiVersion: tekton.dev/v1
+metadata:
+  name: bats-integration
+spec:
+  description: |
+    Test the newly-built ramalama image and layered images on all supported architectures.
+  params:
+  - name: SNAPSHOT
+    description: >-
+      Information about the components included in the current snapshot under test.
+  - name: platforms
+    description: VM platforms on which to run test commands
+    type: array
+    default:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: commands
+    description: Test commands to run
+    type: array
+    default:
+    - make bats
+  - name: git-url
+    description: URL of the Git repository containing pipeline and task definitions
+    default: https://github.com/containers/ramalama.git
+  - name: git-revision
+    description: Revision of the Git repository containing pipeline and task definitions
+    default: main
+  tasks:
+  - name: init
+    params:
+    - name: SNAPSHOT
+      value: $(params.SNAPSHOT)
+    taskRef:
+      resolver: git
+      params:
+      - name: url
+        value: $(params.git-url)
+      - name: revision
+        value: $(params.git-revision)
+      - name: pathInRepo
+        value: .tekton/integration/tasks/init-snapshot.yaml
+  - name: test
+    matrix:
+      params:
+      - name: PLATFORM
+        value:
+        - $(params.platforms)
+      - name: cmd
+        value:
+        - $(params.commands)
+    params:
+    - name: image
+      value: $(tasks.init.results.bats-image)
+    - name: envs
+      value:
+      - RAMALAMA_IMAGE=$(tasks.init.results.ramalama-image)
+    taskRef:
+      resolver: git
+      params:
+      - name: url
+        value: $(params.git-url)
+      - name: revision
+        value: $(params.git-revision)
+      - name: pathInRepo
+        value: .tekton/integration/tasks/test-vm-cmd.yaml

.tekton/integration/tasks/init-snapshot.yaml

@@ -0,0 +1,55 @@
+apiVersion: tekton.dev/v1
+kind: Task
+metadata:
+  name: init-snapshot
+spec:
+  description: Extract information from the SNAPSHOT and make it available as Tekton results
+  params:
+  - name: SNAPSHOT
+    description: >-
+      Information about the components included in the current snapshot under test.
+  results:
+  - name: event-type
+    description: The type of event that triggered the pipeline
+  - name: bats-image
+    description: URI of the bats image included in the snapshot
+  - name: ramalama-image
+    description: URI of the ramalama image included in the snapshot
+  - name: TEST_OUTPUT
+    description: Test result in json format
+  steps:
+  - name: process
+    image: registry.access.redhat.com/ubi10/ubi:latest
+    env:
+    - name: SNAPSHOT
+      value: $(params.SNAPSHOT)
+    - name: EVENT_TYPE
+      valueFrom:
+        fieldRef:
+          fieldPath: metadata.labels['pac.test.appstudio.openshift.io/event-type']
+    - name: RESULTS_EVENT_TYPE_PATH
+      value: $(results.event-type.path)
+    - name: RESULTS_BATS_IMAGE_PATH
+      value: $(results.bats-image.path)
+    - name: RESULTS_RAMALAMA_IMAGE_PATH
+      value: $(results.ramalama-image.path)
+    - name: RESULTS_TEST_OUTPUT_PATH
+      value: $(results.TEST_OUTPUT.path)
+    script: |
+      #!/bin/bash -ex
+      dnf -y install jq
+      echo -n "$EVENT_TYPE" | tee "$RESULTS_EVENT_TYPE_PATH"
+      echo
+      component_image() {
+        TAGSEP=":"
+        if [ "$EVENT_TYPE" == "pull_request" ]; then
+          TAGSEP+="on-pr-"
+        fi
+        jq -j --arg name "$1" --arg tagsep "$TAGSEP" '.components[] | select(.name == $name) | [(.containerImage | split("@")[0]), .source.git.revision] | join($tagsep)' <<< "$SNAPSHOT"
+      }
+      component_image bats | tee "$RESULTS_BATS_IMAGE_PATH"
+      echo
+      component_image ramalama | tee "$RESULTS_RAMALAMA_IMAGE_PATH"
+      echo
+      jq -jnc '{result: "SUCCESS", timestamp: now | todateiso8601, failures: 0, successes: 1, warnings: 0}' | tee "$RESULTS_TEST_OUTPUT_PATH"
+      echo

.tekton/integration/tasks/test-vm-cmd.yaml

@@ -0,0 +1,118 @@
+apiVersion: tekton.dev/v1
+kind: Task
+metadata:
+  name: test-vm-cmd
+spec:
+  description: Run a command in a test VM
+  params:
+  - name: PLATFORM
+    description: The platform of the VM to provision
+  - name: image
+    description: The image to use when setting up the test environment
+  - name: cmd
+    description: The command to run
+  - name: envs
+    description: List of environment variables (NAME=VALUE) to be set in the test environment
+    type: array
+    default: []
+  results:
+  - name: TEST_OUTPUT
+    description: Test result in json format
+  volumes:
+  - name: workdir
+    emptyDir: {}
+  - name: ssh
+    secret:
+      secretName: multi-platform-ssh-$(context.taskRun.name)
+  steps:
+  - name: run-in-vm
+    image: registry.access.redhat.com/ubi10/ubi:latest
+    volumeMounts:
+    - mountPath: /var/workdir
+      name: workdir
+    - mountPath: /ssh
+      name: ssh
+    workingDir: /var/workdir
+    env:
+    - name: TEST_IMAGE
+      value: $(params.image)
+    - name: TEST_CMD
+      value: $(params.cmd)
+    - name: RESULTS_TEST_OUTPUT_PATH
+      value: $(results.TEST_OUTPUT.path)
+    args:
+    - $(params.envs[*])
+    script: |
+      #!/bin/bash -ex
+      log() {
+        echo "[$(date -uIns)]" $*
+      }
+
+      log Install packages
+      dnf -y install openssh-clients rsync jq
+
+      log Prepare connection
+
+      if [ -e "/ssh/error" ]; then
+        log Error provisioning VM
+        cat /ssh/error
+        exit 1
+      fi
+      export SSH_HOST=$(cat /ssh/host)
+
+      mkdir -p ~/.ssh
+      if [ "$SSH_HOST" == "localhost" ] ; then
+        IS_LOCALHOST=true
+        log Localhost detected, running build in cluster
+      elif [ -s "/ssh/otp" ]; then
+        log Fetching OTP token
+        curl --cacert /ssh/otp-ca -d @/ssh/otp $(cat /ssh/otp-server) > ~/.ssh/id_rsa
+        echo >> ~/.ssh/id_rsa
+        chmod 0400 ~/.ssh/id_rsa
+      elif [ -s "/ssh/id_rsa" ]; then
+        log Copying ssh key
+        cp /ssh/id_rsa ~/.ssh
+        chmod 0400 ~/.ssh/id_rsa
+      else
+        log No authentication mechanism found
+        exit 1
+      fi
+
+      mkdir -p scripts
+
+      PODMAN_ENV=()
+      while [ $# -ne 0 ]; do
+        PODMAN_ENV+=("-e" "$1")
+        shift
+      done
+
+      cat > scripts/test.sh <<SCRIPTEOF
+      #!/bin/bash -ex
+      podman run \
+      --userns=keep-id \
+      --security-opt label=disable \
+      --security-opt unmask=/proc/* \
+      --device /dev/net/tun \
+      --device /dev/fuse \
+      ${PODMAN_ENV[*]} \
+      $TEST_IMAGE $TEST_CMD
+      SCRIPTEOF
+      chmod +x scripts/test.sh
+
+      if ! [[ $IS_LOCALHOST ]]; then
+        log VM exec
+        export BUILD_DIR=$(cat /ssh/user-dir)
+        export SSH_ARGS="-o StrictHostKeyChecking=no -o ServerAliveInterval=60 -o ServerAliveCountMax=10"
+        # ssh once before rsync to retrieve the host key
+        ssh $SSH_ARGS "$SSH_HOST" "uname -a"
+        rsync -ra scripts "$SSH_HOST:$BUILD_DIR"
+        ssh $SSH_ARGS "$SSH_HOST" "$BUILD_DIR/scripts/test.sh"
+        log End VM exec
+      else
+        log Local exec
+        scripts/test.sh
+        log End local exec
+      fi
+
+      jq -jnc '{result: "SUCCESS", timestamp: now | todateiso8601, failures: 0, successes: 1, warnings: 0}' | tee "$RESULTS_TEST_OUTPUT_PATH"
+      echo

.tekton/intel-gpu-llama-server/intel-gpu-llama-server-pull-request.yaml

@@ -0,0 +1,47 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-llama-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu-llama-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/intel-gpu-llama-server/intel-gpu-llama-server-push.yaml

@@ -0,0 +1,44 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-llama-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu-llama-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/intel-gpu-rag/intel-gpu-rag-pull-request.yaml

@@ -0,0 +1,47 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-rag-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu-rag:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:on-pr-{{revision}}
+    - GPU=cpu
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/intel-gpu-rag/intel-gpu-rag-push.yaml

@@ -0,0 +1,44 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-rag-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu-rag:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:{{revision}}
+    - GPU=cpu
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/intel-gpu-whisper-server/intel-gpu-whisper-server-pull-request.yaml

@@ -0,0 +1,47 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-whisper-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu-whisper-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/intel-gpu-whisper-server/intel-gpu-whisper-server-push.yaml

@@ -0,0 +1,44 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-whisper-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu-whisper-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/intel-gpu/intel-gpu-pull-request.yaml

@@ -0,0 +1,41 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+  - name: dockerfile
+    value: container-images/intel-gpu/Containerfile
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/intel-gpu/intel-gpu-push.yaml

@@ -0,0 +1,38 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: intel-gpu
+    pipelines.appstudio.openshift.io/type: build
+  name: intel-gpu-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/intel-gpu:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+  - name: dockerfile
+    value: container-images/intel-gpu/Containerfile
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/llama-stack/llama-stack-pull-request.yaml

@@ -0,0 +1,42 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: llama-stack
+    pipelines.appstudio.openshift.io/type: build
+  name: llama-stack-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/llama-stack:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/llama-stack/Containerfile
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/llama-stack/llama-stack-push.yaml

@@ -0,0 +1,39 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: llama-stack
+    pipelines.appstudio.openshift.io/type: build
+  name: llama-stack-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/llama-stack:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/llama-stack/Containerfile
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa-llama-server/musa-llama-server-pull-request.yaml

@@ -0,0 +1,47 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-llama-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa-llama-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/musa:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa-llama-server/musa-llama-server-push.yaml

@@ -0,0 +1,44 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa-llama-server
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-llama-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa-llama-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/musa:{{revision}}
+    - ENTRYPOINT=/usr/bin/llama-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa-rag/musa-rag-pull-request.yaml

@@ -0,0 +1,47 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-rag-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa-rag:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/musa:on-pr-{{revision}}
+    - GPU=musa
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa-rag/musa-rag-push.yaml

@@ -0,0 +1,44 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa-rag
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-rag-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa-rag:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-d160-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.rag
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/musa:{{revision}}
+    - GPU=musa
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa-whisper-server/musa-whisper-server-pull-request.yaml

@@ -0,0 +1,47 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-whisper-server-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa-whisper-server:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:on-pr-{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/musa:on-pr-{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa-whisper-server/musa-whisper-server-push.yaml

@@ -0,0 +1,44 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa-whisper-server
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-whisper-server-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa-whisper-server:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-m2xlarge/amd64
+  - name: dockerfile
+    value: container-images/common/Containerfile.entrypoint
+  - name: parent-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:{{revision}}
+  - name: build-args
+    value:
+    - PARENT=quay.io/redhat-user-workloads/ramalama-tenant/musa:{{revision}}
+    - ENTRYPOINT=/usr/bin/whisper-server.sh
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa/musa-pull-request.yaml

@@ -0,0 +1,41 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+  - name: dockerfile
+    value: container-images/musa/Containerfile
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/musa/musa-push.yaml

@@ -0,0 +1,38 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: musa
+    pipelines.appstudio.openshift.io/type: build
+  name: musa-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/musa:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+  - name: dockerfile
+    value: container-images/musa/Containerfile
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/openvino/openvino-pull-request.yaml

@@ -0,0 +1,41 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: openvino
+    pipelines.appstudio.openshift.io/type: build
+  name: openvino-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/openvino:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+  - name: dockerfile
+    value: container-images/openvino/Containerfile
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/openvino/openvino-push.yaml

@@ -0,0 +1,38 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: openvino
+    pipelines.appstudio.openshift.io/type: build
+  name: openvino-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/openvino:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+  - name: dockerfile
+    value: container-images/openvino/Containerfile
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/pipelines/pull-request-pipeline.yaml

@@ -20,13 +20,11 @@ spec:
     name: output-image
     type: string
   - default: .
-    description: Path to the source code of an application's component from where
-      to build image.
+    description: Path to the source code of an application's component from where to build image.
     name: path-context
     type: string
   - default: Dockerfile
-    description: Path to the Dockerfile inside the context specified by parameter
-      path-context
+    description: Path to the Dockerfile inside the context specified by parameter path-context
     name: dockerfile
     type: string
   - default: "false"
@@ -46,8 +44,7 @@ spec:
     name: prefetch-input
     type: string
   - default: ""
-    description: Image tag expiration time, time values could be something like
-      1h, 2d, 3w for hours, days, and weeks, respectively.
+    description: Image tag expiration time, time values could be something like 1h, 2d, 3w for hours, days, and weeks, respectively.
     name: image-expires-after
   - default: "false"
     description: Build a source image.
@@ -66,14 +63,12 @@ spec:
     name: build-args-file
     type: string
   - default: "false"
-    description: Whether to enable privileged mode, should be used only with remote
-      VMs
+    description: Whether to enable privileged mode, should be used only with remote VMs
     name: privileged-nested
     type: string
   - default:
-    - linux/x86_64
-    description: List of platforms to build the container images on. The available
-      set of values is determined by the configuration of the multi-platform-controller.
+    - linux-c4xlarge/amd64
+    description: List of platforms to build the container images on. The available set of values is determined by the configuration of the multi-platform-controller.
     name: build-platforms
     type: array
   - default: ""
@@ -118,7 +113,7 @@ spec:
       - name: name
         value: init
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-init:0.2@sha256:66e90d31e1386bf516fb548cd3e3f0082b5d0234b8b90dbf9e0d4684b70dbe1a
+        value: quay.io/konflux-ci/tekton-catalog/task-init:0.2@sha256:1d8221c84f91b923d89de50bf16481ea729e3b68ea04a9a7cbe8485ddbb27ee6
       - name: kind
         value: task
       resolver: bundles
@@ -168,7 +163,7 @@ spec:
       - name: name
         value: prefetch-dependencies-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-prefetch-dependencies-oci-ta:0.2@sha256:5e15408f997557153b13d492aeccb51c01923bfbe4fbdf6f1e8695ce1b82f826
+        value: quay.io/konflux-ci/tekton-catalog/task-prefetch-dependencies-oci-ta:0.2@sha256:092491ac0f6e1009d10c58a1319d1029371bf637cc1293cceba53c6da5314ed1
       - name: kind
         value: task
       resolver: bundles
@@ -230,7 +225,7 @@ spec:
       - name: name
         value: buildah-remote-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-buildah-remote-oci-ta:0.4@sha256:ae87472f60dbbf71e4980cd478c92740c145fd9e44acbb9b164a21f1bcd61aa3
+        value: quay.io/konflux-ci/tekton-catalog/task-buildah-remote-oci-ta:0.4@sha256:9e866d4d0489a6ab84ae263db416c9f86d2d6117ef4444f495a0e97388ae3ac0
       - name: kind
         value: task
       resolver: bundles
@@ -259,7 +254,7 @@ spec:
       - name: name
         value: build-image-index
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-build-image-index:0.1@sha256:846dc9975914f31380ec2712fdbac9df3b06c00a9cc7df678315a7f97145efc2
+        value: quay.io/konflux-ci/tekton-catalog/task-build-image-index:0.1@sha256:3499772af90aad0d3935629be6d37dd9292195fb629e6f43ec839c7f545a0faa
       - name: kind
         value: task
       resolver: bundles
@@ -290,8 +285,6 @@ spec:
     params:
     - name: image
       value: $(params.test-image)@$(tasks.wait-for-test-image.results.digest)
-    - name: source-artifact
-      value: $(tasks.prefetch-dependencies.results.SOURCE_ARTIFACT)
     - name: envs
       value:
       - $(params.test-envs[*])
@@ -307,11 +300,13 @@ spec:
   - name: build-source-image
     params:
     - name: BINARY_IMAGE
-      value: $(params.output-image)
+      value: $(tasks.build-image-index.results.IMAGE_URL)
     - name: SOURCE_ARTIFACT
       value: $(tasks.prefetch-dependencies.results.SOURCE_ARTIFACT)
     - name: CACHI2_ARTIFACT
       value: $(tasks.prefetch-dependencies.results.CACHI2_ARTIFACT)
+    - name: BINARY_IMAGE_DIGEST
+      value: $(tasks.build-image-index.results.IMAGE_DIGEST)
     runAfter:
     - build-image-index
     taskRef:
@@ -319,7 +314,7 @@ spec:
       - name: name
         value: source-build-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-source-build-oci-ta:0.2@sha256:b424894fc8e806c12658daa565b835fd2d66e7f7608afc47529eb7b410f030d7
+        value: quay.io/konflux-ci/tekton-catalog/task-source-build-oci-ta:0.3@sha256:b1eb49583b41872b27356fee20d5f0eb6ff7f5cdeacde7ffb39655f031104728
       - name: kind
         value: task
       resolver: bundles
@@ -367,7 +362,7 @@ spec:
       - name: name
         value: clair-scan
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-clair-scan:0.2@sha256:d354939892f3a904223ec080cc3771bd11931085a5d202323ea491ee8e8c5e43
+        value: quay.io/konflux-ci/tekton-catalog/task-clair-scan:0.2@sha256:417f44117f8d87a4a62fea6589b5746612ac61640b454dbd88f74892380411f2
       - name: kind
         value: task
       resolver: bundles
@@ -387,7 +382,7 @@ spec:
       - name: name
         value: ecosystem-cert-preflight-checks
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-ecosystem-cert-preflight-checks:0.2@sha256:b550ff4f0b634512ce5200074be7afd7a5a6c05b783620c626e2a3035cd56448
+        value: quay.io/konflux-ci/tekton-catalog/task-ecosystem-cert-preflight-checks:0.2@sha256:f99d2bdb02f13223d494077a2cde31418d09369f33c02134a8e7e5fad2f61eda
       - name: kind
         value: task
       resolver: bundles
@@ -413,7 +408,7 @@ spec:
       - name: name
         value: sast-snyk-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-snyk-check-oci-ta:0.4@sha256:e61f541189b30d14292ef8df36ccaf13f7feb2378fed5f74cb6293b3e79eb687
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-snyk-check-oci-ta:0.4@sha256:fe5e5ba3a72632cd505910de2eacd62c9d11ed570c325173188f8d568ac60771
       - name: kind
         value: task
       resolver: bundles
@@ -435,7 +430,7 @@ spec:
       - name: name
         value: clamav-scan
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-clamav-scan:0.2@sha256:9cab95ac9e833d77a63c079893258b73b8d5a298d93aaf9bdd6722471bc2f338
+        value: quay.io/konflux-ci/tekton-catalog/task-clamav-scan:0.2@sha256:7749146f7e4fe530846f1b15c9366178ec9f44776ef1922a60d3e7e2b8c6426b
       - name: kind
         value: task
       resolver: bundles
@@ -480,7 +475,7 @@ spec:
       - name: name
         value: sast-coverity-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-coverity-check-oci-ta:0.3@sha256:f81ade665c725616b918356c8c2fb2d4ed972e822a1a3181933cd0ada728a231
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-coverity-check-oci-ta:0.3@sha256:f9ca942208dc2e63b479384ccc56a611cc793397ecc837637b5b9f89c2ecbefe
       - name: kind
         value: task
       resolver: bundles
@@ -527,7 +522,7 @@ spec:
       - name: name
         value: sast-shell-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-shell-check-oci-ta:0.1@sha256:808bcaf75271db6a999f53fdefb973a385add94a277d37fbd3df68f8ac7dfaa3
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-shell-check-oci-ta:0.1@sha256:bf7bdde00b7212f730c1356672290af6f38d070da2c8a316987b5c32fd49e0b9
       - name: kind
         value: task
       resolver: bundles
@@ -553,7 +548,7 @@ spec:
       - name: name
         value: sast-unicode-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-unicode-check-oci-ta:0.2@sha256:24ad71fde435fc25abba2c4c550beb088b1530f738d3c377e2f635b5f320d57b
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-unicode-check-oci-ta:0.3@sha256:a2bde66f6b4164620298c7d709b8f08515409404000fa1dc2260d2508b135651
       - name: kind
         value: task
       resolver: bundles
@@ -598,7 +593,7 @@ spec:
       - name: name
         value: push-dockerfile-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-push-dockerfile-oci-ta:0.1@sha256:5d8013b6a27bbc5e4ff261144616268f28417ed0950d583ef36349fcd59d3d3d
+        value: quay.io/konflux-ci/tekton-catalog/task-push-dockerfile-oci-ta:0.1@sha256:8c75c4a747e635e5f3e12266a3bb6e5d3132bf54e37eaa53d505f89897dd8eca
       - name: kind
         value: task
       resolver: bundles
@@ -634,7 +629,7 @@ spec:
       - name: name
         value: show-sbom
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-show-sbom:0.1@sha256:1b1df4da95966d08ac6a5b8198710e09e68b5c2cdc707c37d9d19769e65884b2
+        value: quay.io/konflux-ci/tekton-catalog/task-show-sbom:0.1@sha256:86c069cac0a669797e8049faa8aa4088e70ff7fcd579d5bdc37626a9e0488a05
       - name: kind
         value: task
       resolver: bundles

.tekton/pipelines/push-pipeline.yaml

@@ -20,13 +20,11 @@ spec:
     name: output-image
     type: string
   - default: .
-    description: Path to the source code of an application's component from where
-      to build image.
+    description: Path to the source code of an application's component from where to build image.
     name: path-context
     type: string
   - default: Dockerfile
-    description: Path to the Dockerfile inside the context specified by parameter
-      path-context
+    description: Path to the Dockerfile inside the context specified by parameter path-context
     name: dockerfile
     type: string
   - default: "false"
@@ -46,8 +44,7 @@ spec:
     name: prefetch-input
     type: string
   - default: ""
-    description: Image tag expiration time, time values could be something like
-      1h, 2d, 3w for hours, days, and weeks, respectively.
+    description: Image tag expiration time, time values could be something like 1h, 2d, 3w for hours, days, and weeks, respectively.
     name: image-expires-after
   - default: "false"
     description: Build a source image.
@@ -66,14 +63,12 @@ spec:
     name: build-args-file
     type: string
   - default: "false"
-    description: Whether to enable privileged mode, should be used only with remote
-      VMs
+    description: Whether to enable privileged mode, should be used only with remote VMs
     name: privileged-nested
     type: string
   - default:
-    - linux/x86_64
-    description: List of platforms to build the container images on. The available
-      set of values is determined by the configuration of the multi-platform-controller.
+    - linux-c4xlarge/amd64
+    description: List of platforms to build the container images on. The available set of values is determined by the configuration of the multi-platform-controller.
     name: build-platforms
     type: array
   - default: ""
@@ -118,7 +113,7 @@ spec:
       - name: name
         value: init
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-init:0.2@sha256:66e90d31e1386bf516fb548cd3e3f0082b5d0234b8b90dbf9e0d4684b70dbe1a
+        value: quay.io/konflux-ci/tekton-catalog/task-init:0.2@sha256:1d8221c84f91b923d89de50bf16481ea729e3b68ea04a9a7cbe8485ddbb27ee6
       - name: kind
         value: task
       resolver: bundles
@@ -168,7 +163,7 @@ spec:
       - name: name
         value: prefetch-dependencies-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-prefetch-dependencies-oci-ta:0.2@sha256:5e15408f997557153b13d492aeccb51c01923bfbe4fbdf6f1e8695ce1b82f826
+        value: quay.io/konflux-ci/tekton-catalog/task-prefetch-dependencies-oci-ta:0.2@sha256:092491ac0f6e1009d10c58a1319d1029371bf637cc1293cceba53c6da5314ed1
       - name: kind
         value: task
       resolver: bundles
@@ -230,7 +225,7 @@ spec:
       - name: name
         value: buildah-remote-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-buildah-remote-oci-ta:0.4@sha256:ae87472f60dbbf71e4980cd478c92740c145fd9e44acbb9b164a21f1bcd61aa3
+        value: quay.io/konflux-ci/tekton-catalog/task-buildah-remote-oci-ta:0.4@sha256:9e866d4d0489a6ab84ae263db416c9f86d2d6117ef4444f495a0e97388ae3ac0
       - name: kind
         value: task
       resolver: bundles
@@ -259,7 +254,7 @@ spec:
       - name: name
         value: build-image-index
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-build-image-index:0.1@sha256:846dc9975914f31380ec2712fdbac9df3b06c00a9cc7df678315a7f97145efc2
+        value: quay.io/konflux-ci/tekton-catalog/task-build-image-index:0.1@sha256:3499772af90aad0d3935629be6d37dd9292195fb629e6f43ec839c7f545a0faa
       - name: kind
         value: task
       resolver: bundles
@@ -290,8 +285,6 @@ spec:
     params:
     - name: image
       value: $(params.test-image)@$(tasks.wait-for-test-image.results.digest)
-    - name: source-artifact
-      value: $(tasks.prefetch-dependencies.results.SOURCE_ARTIFACT)
     - name: envs
       value:
       - $(params.test-envs[*])
@@ -307,11 +300,13 @@ spec:
   - name: build-source-image
     params:
     - name: BINARY_IMAGE
-      value: $(params.output-image)
+      value: $(tasks.build-image-index.results.IMAGE_URL)
     - name: SOURCE_ARTIFACT
       value: $(tasks.prefetch-dependencies.results.SOURCE_ARTIFACT)
     - name: CACHI2_ARTIFACT
       value: $(tasks.prefetch-dependencies.results.CACHI2_ARTIFACT)
+    - name: BINARY_IMAGE_DIGEST
+      value: $(tasks.build-image-index.results.IMAGE_DIGEST)
     runAfter:
     - build-image-index
     taskRef:
@@ -319,7 +314,7 @@ spec:
       - name: name
         value: source-build-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-source-build-oci-ta:0.2@sha256:b424894fc8e806c12658daa565b835fd2d66e7f7608afc47529eb7b410f030d7
+        value: quay.io/konflux-ci/tekton-catalog/task-source-build-oci-ta:0.3@sha256:b1eb49583b41872b27356fee20d5f0eb6ff7f5cdeacde7ffb39655f031104728
       - name: kind
         value: task
       resolver: bundles
@@ -367,7 +362,7 @@ spec:
       - name: name
         value: clair-scan
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-clair-scan:0.2@sha256:d354939892f3a904223ec080cc3771bd11931085a5d202323ea491ee8e8c5e43
+        value: quay.io/konflux-ci/tekton-catalog/task-clair-scan:0.2@sha256:417f44117f8d87a4a62fea6589b5746612ac61640b454dbd88f74892380411f2
       - name: kind
         value: task
       resolver: bundles
@@ -387,7 +382,7 @@ spec:
       - name: name
         value: ecosystem-cert-preflight-checks
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-ecosystem-cert-preflight-checks:0.2@sha256:b550ff4f0b634512ce5200074be7afd7a5a6c05b783620c626e2a3035cd56448
+        value: quay.io/konflux-ci/tekton-catalog/task-ecosystem-cert-preflight-checks:0.2@sha256:f99d2bdb02f13223d494077a2cde31418d09369f33c02134a8e7e5fad2f61eda
       - name: kind
         value: task
       resolver: bundles
@@ -413,7 +408,7 @@ spec:
       - name: name
         value: sast-snyk-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-snyk-check-oci-ta:0.4@sha256:e61f541189b30d14292ef8df36ccaf13f7feb2378fed5f74cb6293b3e79eb687
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-snyk-check-oci-ta:0.4@sha256:fe5e5ba3a72632cd505910de2eacd62c9d11ed570c325173188f8d568ac60771
       - name: kind
         value: task
       resolver: bundles
@@ -435,7 +430,7 @@ spec:
       - name: name
         value: clamav-scan
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-clamav-scan:0.2@sha256:9cab95ac9e833d77a63c079893258b73b8d5a298d93aaf9bdd6722471bc2f338
+        value: quay.io/konflux-ci/tekton-catalog/task-clamav-scan:0.2@sha256:7749146f7e4fe530846f1b15c9366178ec9f44776ef1922a60d3e7e2b8c6426b
       - name: kind
         value: task
       resolver: bundles
@@ -480,7 +475,7 @@ spec:
       - name: name
         value: sast-coverity-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-coverity-check-oci-ta:0.3@sha256:f81ade665c725616b918356c8c2fb2d4ed972e822a1a3181933cd0ada728a231
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-coverity-check-oci-ta:0.3@sha256:f9ca942208dc2e63b479384ccc56a611cc793397ecc837637b5b9f89c2ecbefe
       - name: kind
         value: task
       resolver: bundles
@@ -527,7 +522,7 @@ spec:
       - name: name
         value: sast-shell-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-shell-check-oci-ta:0.1@sha256:808bcaf75271db6a999f53fdefb973a385add94a277d37fbd3df68f8ac7dfaa3
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-shell-check-oci-ta:0.1@sha256:bf7bdde00b7212f730c1356672290af6f38d070da2c8a316987b5c32fd49e0b9
       - name: kind
         value: task
       resolver: bundles
@@ -553,7 +548,7 @@ spec:
       - name: name
         value: sast-unicode-check-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-sast-unicode-check-oci-ta:0.2@sha256:24ad71fde435fc25abba2c4c550beb088b1530f738d3c377e2f635b5f320d57b
+        value: quay.io/konflux-ci/tekton-catalog/task-sast-unicode-check-oci-ta:0.3@sha256:a2bde66f6b4164620298c7d709b8f08515409404000fa1dc2260d2508b135651
       - name: kind
         value: task
       resolver: bundles
@@ -598,7 +593,7 @@ spec:
       - name: name
         value: push-dockerfile-oci-ta
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-push-dockerfile-oci-ta:0.1@sha256:5d8013b6a27bbc5e4ff261144616268f28417ed0950d583ef36349fcd59d3d3d
+        value: quay.io/konflux-ci/tekton-catalog/task-push-dockerfile-oci-ta:0.1@sha256:8c75c4a747e635e5f3e12266a3bb6e5d3132bf54e37eaa53d505f89897dd8eca
       - name: kind
         value: task
       resolver: bundles
@@ -634,7 +629,7 @@ spec:
       - name: name
         value: show-sbom
       - name: bundle
-        value: quay.io/konflux-ci/tekton-catalog/task-show-sbom:0.1@sha256:1b1df4da95966d08ac6a5b8198710e09e68b5c2cdc707c37d9d19769e65884b2
+        value: quay.io/konflux-ci/tekton-catalog/task-show-sbom:0.1@sha256:86c069cac0a669797e8049faa8aa4088e70ff7fcd579d5bdc37626a9e0488a05
       - name: kind
         value: task
       resolver: bundles

.tekton/ramalama-cli/ramalama-cli-pull-request.yaml

@@ -0,0 +1,42 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/pull_request_number: '{{pull_request_number}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "true"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: ramalama-cli
+    pipelines.appstudio.openshift.io/type: build
+  name: ramalama-cli-on-pull-request
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/ramalama-cli:on-pr-{{revision}}
+  - name: image-expires-after
+    value: 5d
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/ramalama-cli/Containerfile
+  pipelineRef:
+    name: pull-request-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/ramalama-cli/ramalama-cli-push.yaml

@@ -0,0 +1,39 @@
+apiVersion: tekton.dev/v1
+kind: PipelineRun
+metadata:
+  annotations:
+    build.appstudio.openshift.io/repo: https://github.com/containers/ramalama?rev={{revision}}
+    build.appstudio.redhat.com/commit_sha: '{{revision}}'
+    build.appstudio.redhat.com/target_branch: '{{target_branch}}'
+    pipelinesascode.tekton.dev/cancel-in-progress: "false"
+    pipelinesascode.tekton.dev/max-keep-runs: "3"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
+  labels:
+    appstudio.openshift.io/application: ramalama
+    appstudio.openshift.io/component: ramalama-cli
+    pipelines.appstudio.openshift.io/type: build
+  name: ramalama-cli-on-push
+  namespace: ramalama-tenant
+spec:
+  params:
+  - name: git-url
+    value: '{{source_url}}'
+  - name: revision
+    value: '{{revision}}'
+  - name: output-image
+    value: quay.io/redhat-user-workloads/ramalama-tenant/ramalama-cli:{{revision}}
+  - name: build-platforms
+    value:
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
+  - name: dockerfile
+    value: container-images/ramalama-cli/Containerfile
+  pipelineRef:
+    name: push-pipeline
+  timeouts:
+    pipeline: 6h
+  workspaces:
+  - name: git-auth
+    secret:
+      secretName: '{{ git_auth_secret }}'

.tekton/ramalama-llama-server/ramalama-llama-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama-llama-server
@@ -28,7 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/ramalama-llama-server/ramalama-llama-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama-llama-server
@@ -25,7 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/ramalama-llama-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/ramalama-rag/ramalama-rag-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama-rag
@@ -28,7 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/ramalama-rag/ramalama-rag-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama-rag
@@ -25,7 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/ramalama-rag:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
+    - linux-d160-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/ramalama-whisper-server/ramalama-whisper-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama-whisper-server
@@ -28,7 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/ramalama-whisper-server/ramalama-whisper-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama-whisper-server
@@ -25,7 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/ramalama-whisper-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
+    - linux-m2xlarge/arm64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/ramalama/ramalama-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama
@@ -28,8 +28,8 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
-    - linux/arm64
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
   - name: dockerfile
     value: container-images/ramalama/Containerfile
   - name: test-image
@@ -40,7 +40,6 @@ spec:
   - name: test-commands
     value:
     - make validate
-    - make bats-nocontainer
     - make unit-tests
     - make cov-tests
   pipelineRef:

.tekton/ramalama/ramalama-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: ramalama
@@ -25,8 +25,8 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/ramalama:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
-    - linux/arm64
+    - linux-c4xlarge/amd64
+    - linux-c4xlarge/arm64
   - name: dockerfile
     value: container-images/ramalama/Containerfile
   - name: test-image
@@ -37,7 +37,6 @@ spec:
   - name: test-commands
     value:
     - make validate
-    - make bats-nocontainer
     - make unit-tests
     - make cov-tests
   pipelineRef:

.tekton/rocm-llama-server/rocm-llama-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-llama-server
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm-llama-server/rocm-llama-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-llama-server
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm-llama-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm-rag/rocm-rag-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-rag
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/rocm-rag/rocm-rag-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-rag
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm-rag:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/rocm-ubi-llama-server/rocm-ubi-llama-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi-llama-server
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm-ubi-llama-server/rocm-ubi-llama-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi-llama-server
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm-ubi-llama-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm-ubi-rag/rocm-ubi-rag-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi-rag
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/rocm-ubi-rag/rocm-ubi-rag-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi-rag
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm-ubi-rag:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-d160-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.rag
   - name: parent-image

.tekton/rocm-ubi-whisper-server/rocm-ubi-whisper-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi-whisper-server
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm-ubi-whisper-server/rocm-ubi-whisper-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi-whisper-server
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm-ubi-whisper-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm-ubi/rocm-ubi-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-fast/amd64
   - name: dockerfile
     value: container-images/rocm-ubi/Containerfile
   pipelineRef:

.tekton/rocm-ubi/rocm-ubi-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-ubi
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm-ubi:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-fast/amd64
   - name: dockerfile
     value: container-images/rocm-ubi/Containerfile
   pipelineRef:

.tekton/rocm-whisper-server/rocm-whisper-server-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-whisper-server
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm-whisper-server/rocm-whisper-server-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm-whisper-server
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm-whisper-server:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-m2xlarge/amd64
   - name: dockerfile
     value: container-images/common/Containerfile.entrypoint
   - name: parent-image

.tekton/rocm/rocm-pull-request.yaml

@@ -8,8 +8,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "true"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "pull_request" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "pull_request" && target_branch == "main" && body.action != "ready_for_review"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm
@@ -28,7 +28,7 @@ spec:
     value: 5d
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-fast/amd64
   - name: dockerfile
     value: container-images/rocm/Containerfile
   pipelineRef:

.tekton/rocm/rocm-push.yaml

@@ -7,8 +7,8 @@ metadata:
     build.appstudio.redhat.com/target_branch: '{{target_branch}}'
     pipelinesascode.tekton.dev/cancel-in-progress: "false"
     pipelinesascode.tekton.dev/max-keep-runs: "3"
-    pipelinesascode.tekton.dev/on-cel-expression: event == "push" && target_branch
-      == "main"
+    pipelinesascode.tekton.dev/on-cel-expression: >-
+      event == "push" && target_branch == "main"
   labels:
     appstudio.openshift.io/application: ramalama
     appstudio.openshift.io/component: rocm
@@ -25,7 +25,7 @@ spec:
     value: quay.io/redhat-user-workloads/ramalama-tenant/rocm:{{revision}}
   - name: build-platforms
     value:
-    - linux/x86_64
+    - linux-fast/amd64
   - name: dockerfile
     value: container-images/rocm/Containerfile
   pipelineRef:

.tekton/tasks/test-cmd.yaml

@@ -7,58 +7,29 @@ spec:
   params:
   - name: image
     description: The image to use when setting up the test environment.
-  - name: source-artifact
-    description: The Trusted Artifact URI pointing to the artifact with the application source code.
   - name: cmd
     description: The command to run.
   - name: envs
     description: List of environment variables (NAME=VALUE) to be set in the test environment.
     type: array
     default: []
-  volumes:
-  - name: workdir
-    emptyDir: {}
-  stepTemplate:
-    volumeMounts:
-    - mountPath: /var/workdir
-      name: workdir
+  steps:
+  - name: run
+    image: $(params.image)
     computeResources:
       limits:
         memory: 4Gi
       requests:
         cpu: "1"
         memory: 1Gi
-  steps:
-  - name: use-trusted-artifact
-    image: quay.io/konflux-ci/build-trusted-artifacts:latest@sha256:4689f88dd253bd1feebf57f1a76a5a751880f739000719cd662bbdc76990a7fd
-    args:
-    - use
-    - $(params.source-artifact)=/var/workdir/source
-  - name: set-env
-    image: $(params.image)
-    workingDir: /var/workdir/source
-    args:
-    - $(params.envs[*])
-    script: |
-      #!/bin/bash -e
-      rm -f .bashenv
-      while [ $# -ne 0 ]; do
-        echo "$1" >> .bashenv
-        shift
-      done
-  - name: run
-    image: $(params.image)
     securityContext:
       capabilities:
         add:
         - SETFCAP
-    workingDir: /var/workdir/source
-    env:
-    - name: BASH_ENV
-      value: .bashenv
     command:
     - /usr/bin/entrypoint.sh
     args:
+    - $(params.envs[*])
     - /bin/bash
     - -ex
     - -c

Makefile

@@ -168,7 +168,7 @@ bats-image:
 	podman inspect $(BATS_IMAGE) &> /dev/null || \
 		podman build -t $(BATS_IMAGE) -f container-images/bats/Containerfile .
 
-bats-in-container: extra-opts = --security-opt unmask=/proc/* --device /dev/net/tun
+bats-in-container: extra-opts = --security-opt unmask=/proc/* --device /dev/net/tun --device /dev/fuse
 
 %-in-container: bats-image
 	podman run -it --rm \

README.md

@@ -2,14 +2,12 @@
 <img src="https://github.com/user-attachments/assets/1a338ecf-dc84-4495-8c70-16882955da47" width=50%>
 </p>
 
-[RamaLama](https://ramalama.ai) is an open-source tool that simplifies the local use and serving of AI models for inference from any source through the familiar approach of containers.
+[RamaLama](https://ramalama.ai) strives to make working with AI simple, straightforward, and familiar by using OCI containers.
 <br>
 <br>
 
 ## Description
-RamaLama strives to make working with AI simple, straightforward, and familiar by using OCI containers.
-
-RamaLama is an open-source tool that simplifies the local use and serving of AI models for inference from any source through the familiar approach of containers. Using a container engine like Podman, engineers can use container-centric development patterns and benefits to extend to AI use cases.
+RamaLama is an open-source tool that simplifies the local use and serving of AI models for inference from any source through the familiar approach of containers. It allows engineers to use container-centric development patterns and benefits to extend to AI use cases.
 
 RamaLama eliminates the need to configure the host system by instead pulling a container image specific to the GPUs discovered on the host system, and allowing you to work with various models and platforms.
 
@@ -23,6 +21,25 @@ RamaLama eliminates the need to configure the host system by instead pulling a c
 - Interact with models via REST API or as a chatbot.
 <br>
 
+## Install
+### Install on Fedora
+RamaLama is available in [Fedora](https://fedoraproject.org/) and later. To install it, run:
+```
+sudo dnf install python3-ramalama
+```
+
+### Install via PyPI
+RamaLama is available via PyPI at [https://pypi.org/project/ramalama](https://pypi.org/project/ramalama)
+```
+pip install ramalama
+```
+
+### Install script (Linux and macOS)
+Install RamaLama by running:
+```
+curl -fsSL https://ramalama.ai/install.sh | bash
+```
+
 ## Accelerated images
 
 | Accelerator                       | Image                      |
@@ -103,30 +120,6 @@ pip install mlx-lm
 ramalama --runtime=mlx serve hf://mlx-community/Unsloth-Phi-4-4bit
 ```
 
-## Install
-### Install on Fedora
-RamaLama is available in [Fedora 40](https://fedoraproject.org/) and later. To install it, run:
-```
-sudo dnf install python3-ramalama
-```
-
-### Install via PyPi
-RamaLama is available via PyPi at [https://pypi.org/project/ramalama](https://pypi.org/project/ramalama)
-```
-pip install ramalama
-```
-
-### Install via Homebrew
-```
-brew install ramalama
-```
-
-### Install script (Linux and macOS)
-Install RamaLama by running:
-```
-curl -fsSL https://ramalama.ai/install.sh | bash
-```
-
 #### Default Container Engine
 When both Podman and Docker are installed, RamaLama defaults to Podman. The `RAMALAMA_CONTAINER_ENGINE=docker` environment variable can override this behaviour. When neither are installed, RamaLama will attempt to run the model with software on the local system.
 <br>
@@ -229,7 +222,7 @@ $ cat /usr/share/ramalama/shortnames.conf
 	<br>
 
 	```
-	$ ramalama bench granite-moe3
+	$ ramalama bench granite3-moe
 	```
 </details>
 
@@ -836,7 +829,7 @@ $ cat /usr/share/ramalama/shortnames.conf
 
 	Perplexity measures how well the model can predict the next token with lower values being better
 	```
-	$ ramalama perplexity granite-moe3
+	$ ramalama perplexity granite3-moe
 	```
 </details>
 

container-images/bats/Containerfile

@@ -1,12 +1,11 @@
 FROM quay.io/fedora/fedora:42
 
 ENV HOME=/tmp \
-    XDG_RUNTIME_DIR=/tmp \
-    STORAGE_DRIVER=vfs
+    XDG_RUNTIME_DIR=/tmp
 WORKDIR /src
 ENTRYPOINT ["/usr/bin/entrypoint.sh"]
 
-RUN dnf -y install make bats jq iproute podman openssl httpd-tools \
+RUN dnf -y install make bats jq iproute podman openssl httpd-tools diffutils \
     python3-huggingface-hub \
     $([ $(uname -m) == "x86_64" ] && echo ollama) \
     # for validate and unit-tests
@@ -26,4 +25,6 @@ RUN git clone --depth=1 https://github.com/ggml-org/llama.cpp && \
 
 COPY container-images/bats/entrypoint.sh /usr/bin
 COPY container-images/bats/containers.conf /etc/containers
+COPY . /src
+RUN chmod -R a+rw /src
 RUN chmod a+rw /etc/subuid /etc/subgid

container-images/bats/entrypoint.sh

@@ -3,6 +3,16 @@
 echo "$(id -un):10000:2000" > /etc/subuid
 echo "$(id -un):10000:2000" > /etc/subgid
 
+while [ $# -gt 0 ]; do
+    if [[ "$1" =~ = ]]; then
+        # shellcheck disable=SC2163
+        export "$1"
+        shift
+    else
+        break
+    fi
+done
+
 if [ $# -gt 0 ]; then
     exec "$@"
 else

container-images/llama-stack/Containerfile

@@ -1,6 +1,6 @@
 FROM quay.io/fedora/fedora:42
 
-ARG RAMALAMA_STACK_VERSION=0.2.4
+ARG RAMALAMA_STACK_VERSION=0.2.5
 
 # hack that should be removed when the following bug is addressed
 # https://github.com/containers/ramalama-stack/issues/53

container-images/openvino/Containerfile

@@ -1,4 +1,4 @@
-FROM registry.access.redhat.com/ubi9/ubi:9.6-1751445649
+FROM registry.access.redhat.com/ubi9/ubi:9.6-1752625787
 
 # Install Python development dependencies
 RUN dnf install -y python3-devel wget compat-openssl11 python3-jinja2 python3-markupsafe

container-images/ramalama-vllm/Containerfile

@@ -0,0 +1,17 @@
+FROM quay.io/ramalama/ramalama
+
+ENV PATH="/root/.local/bin:$PATH"
+ENV VIRTUAL_ENV="/opt/venv"
+ENV UV_PYTHON_INSTALL_DIR="/opt/uv/python"
+ENV PATH="$VIRTUAL_ENV/bin:$PATH"
+
+ENV UV_HTTP_TIMEOUT=500
+
+ENV UV_INDEX_STRATEGY="unsafe-best-match"
+ENV UV_LINK_MODE="copy"
+
+COPY . /src/ramalama
+WORKDIR /src/ramalama
+RUN container-images/scripts/build-vllm.sh
+WORKDIR /
+

container-images/rocm-ubi/Containerfile

@@ -1,4 +1,4 @@
-FROM registry.access.redhat.com/ubi9/ubi:9.6-1751445649
+FROM registry.access.redhat.com/ubi9/ubi:9.6-1752625787
 
 COPY container-images/rocm-ubi/amdgpu.repo /etc/yum.repos.d/
 COPY container-images/rocm-ubi/rocm.repo /etc/yum.repos.d/

container-images/scripts/build-vllm.sh

@@ -0,0 +1,90 @@
+#!/bin/bash
+
+available() {
+  command -v "$1" >/dev/null
+}
+
+install_deps() {
+  set -eux -o pipefail
+
+  if available dnf; then
+    dnf install -y git curl wget ca-certificates gcc gcc-c++ \
+      gperftools-libs numactl-devel ffmpeg libSM libXext mesa-libGL jq lsof \
+      vim numactl
+    dnf -y clean all
+    rm -rf /var/cache/*dnf*
+  elif available apt-get; then
+    apt-get update -y
+    apt-get install -y --no-install-recommends git curl wget ca-certificates \
+      gcc g++ libtcmalloc-minimal4 libnuma-dev ffmpeg libsm6 libxext6 libgl1 \
+      jq lsof vim numactl
+    rm -rf /var/lib/apt/lists/*
+  fi
+
+  curl -LsSf https://astral.sh/uv/0.7.21/install.sh | bash
+}
+
+preload_and_ulimit() {
+  local ld_preload_file="libtcmalloc_minimal.so.4"
+  local ld_preload_file_1="/usr/lib/$arch-linux-gnu/$ld_preload_file"
+  local ld_preload_file_2="/usr/lib64/$ld_preload_file"
+  if [ -e "$ld_preload_file_1" ]; then
+    ld_preload_file="$ld_preload_file_1"
+  elif [ -e "$ld_preload_file_2" ]; then
+    ld_preload_file="$ld_preload_file_2"
+  fi
+
+  if [ -e "$ld_preload_file" ]; then
+    echo "LD_PRELOAD=$ld_preload_file" >> /etc/environment
+  fi
+
+  echo 'ulimit -c 0' >> ~/.bashrc
+}
+
+pip_install() {
+  local url="https://download.pytorch.org/whl/cpu"
+  uv pip install -v -r "$1" --extra-index-url $url
+}
+
+git_clone_specific_commit() {
+  local repo="${vllm_url##*/}"
+  git init "$repo"
+  cd "$repo"
+  git remote add origin "$vllm_url"
+  git fetch --depth 1 origin $commit
+  git reset --hard $commit
+}
+
+main() {
+  set -eux -o pipefail
+
+  install_deps
+
+  local arch
+  arch=$(uname -m)
+  preload_and_ulimit
+
+  uv venv --python 3.12 --seed "$VIRTUAL_ENV"
+  uv pip install --upgrade pip
+
+  local vllm_url="https://github.com/vllm-project/vllm"
+  local commit="ac9fb732a5c0b8e671f8c91be8b40148282bb14a"
+  git_clone_specific_commit
+  if [ "$arch" == "x86_64" ]; then
+    export VLLM_CPU_DISABLE_AVX512="0"
+    export VLLM_CPU_AVX512BF16="0"
+    export VLLM_CPU_AVX512VNNI="0"
+  elif [ "$arch" == "aarch64" ]; then
+    export VLLM_CPU_DISABLE_AVX512="true"
+  fi
+
+  pip_install requirements/cpu-build.txt
+  pip_install requirements/cpu.txt
+
+  MAX_JOBS=2 VLLM_TARGET_DEVICE=cpu python3 setup.py install
+  cd -
+  rm -rf vllm /root/.cache
+}
+
+main "$@"
+

container-images/scripts/build_llama_and_whisper.sh

@@ -99,10 +99,10 @@ is_rhel_based() { # doesn't include openEuler
 dnf_install_mesa() {
   if [ "${ID}" = "fedora" ]; then
     dnf copr enable -y slp/mesa-libkrun-vulkan
-    dnf install -y mesa-vulkan-drivers-25.0.7-100.fc42 "${vulkan_rpms[@]}"
+    dnf install -y mesa-vulkan-drivers-25.0.7-100.fc42 virglrenderer "${vulkan_rpms[@]}"
     dnf versionlock add mesa-vulkan-drivers-25.0.7-100.fc42
   else
-    dnf install -y mesa-vulkan-drivers "${vulkan_rpms[@]}"
+    dnf install -y mesa-vulkan-drivers virglrenderer "${vulkan_rpms[@]}"
   fi
 
   rm_non_ubi_repos
@@ -282,18 +282,22 @@ clone_and_build_llama_cpp() {
 }
 
 install_ramalama() {
-  $PYTHON -m pip install . --prefix="$1"
+  if [ -e "pyproject.toml" ]; then
+    $PYTHON -m pip install . --prefix="$1"
+  fi
 }
 
 install_entrypoints() {
-  install -d "$install_prefix"/bin
-  install -m 755 \
-    container-images/scripts/llama-server.sh \
-    container-images/scripts/whisper-server.sh \
-    container-images/scripts/build_rag.sh \
-    container-images/scripts/doc2rag \
-    container-images/scripts/rag_framework \
-    "$install_prefix"/bin
+  if [ -e "container-images" ]; then
+    install -d "$install_prefix"/bin
+    install -m 755 \
+      container-images/scripts/llama-server.sh \
+      container-images/scripts/whisper-server.sh \
+      container-images/scripts/build_rag.sh \
+      container-images/scripts/doc2rag \
+      container-images/scripts/rag_framework \
+      "$install_prefix"/bin
+  fi
 }
 
 main() {

container-images/scripts/build_rag.sh

@@ -40,7 +40,18 @@ update_python() {
 }
 
 docling() {
-    ${python} -m pip install --prefix=/usr docling docling-core accelerate --extra-index-url https://download.pytorch.org/whl/"$1"
+    case $1 in
+        cuda)
+            PYTORCH_DIR="cu128"
+            ;;
+        rocm)
+            PYTORCH_DIR="rocm6.3"
+            ;;
+        *)
+            PYTORCH_DIR="cpu"
+            ;;
+    esac
+    ${python} -m pip install --prefix=/usr docling docling-core accelerate --extra-index-url "https://download.pytorch.org/whl/$PYTORCH_DIR"
     # Preloads models (assumes its installed from container_build.sh)
     doc2rag load
 }
@@ -51,6 +62,8 @@ rag() {
 }
 
 to_gguf() {
+    # required to build under GCC 15 until a new release is available, see https://github.com/google/sentencepiece/issues/1108 for details
+    export CXXFLAGS="-include cstdint"
     ${python} -m pip install --prefix=/usr "numpy~=1.26.4" "sentencepiece~=0.2.0" "transformers>=4.45.1,<5.0.0" git+https://github.com/ggml-org/llama.cpp#subdirectory=gguf-py "protobuf>=4.21.0,<5.0.0"
 }
 
@@ -60,6 +73,9 @@ main() {
     # shellcheck disable=SC1091
     source /etc/os-release
 
+    # caching in a container build is unhelpful, and can cause errors
+    export PIP_NO_CACHE_DIR=1
+
     local arch
     arch="$(uname -m)"
     local gpu="${1-cpu}"
@@ -67,18 +83,14 @@ main() {
     python=$(python_version)
     local pkgs
     if available dnf; then
-        pkgs=("git-core" "gcc" "gcc-c++")
+        pkgs=("git-core" "gcc" "gcc-c++" "cmake")
     else
-        pkgs=("git" "gcc" "g++")
+        pkgs=("git" "gcc" "g++" "cmake")
     fi
     if [ "${gpu}" = "cuda" ]; then
         pkgs+=("libcudnn9-devel-cuda-12" "libcusparselt0" "cuda-cupti-12-*")
     fi
 
-    if [[ "$ID" = "fedora" && "$VERSION_ID" -ge 42 ]] ; then
-        pkgs+=("python3-sentencepiece-0.2.0")
-    fi
-
     update_python
     to_gguf
 

container-images/scripts/doc2rag

@@ -1,8 +1,10 @@
 #!/usr/bin/env python3
 
 import argparse
+import errno
 import hashlib
 import os
+import os.path
 import sys
 import uuid
 
@@ -23,45 +25,46 @@ COLLECTION_NAME = "rag"
 class Converter:
     """A Class designed to handle all document conversions using Docling"""
 
-    def __init__(self, output, targets, ocr):
+    def __init__(self, args):
         # Docling Setup (Turn off OCR (image processing) for drastically reduced RAM usage and big speed increase)
         pipeline_options = PdfPipelineOptions()
-        pipeline_options.do_ocr = ocr
+        pipeline_options.do_ocr = args.ocr
+        self.sources = []
+        for source in args.sources:
+            self.add(source)
+        self.output = args.output
+        self.format = args.format
         self.doc_converter = DocumentConverter(
             format_options={InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)}
         )
-        self.targets = []
-        for target in targets:
-            self.add(target)
-        self.output = output
-        self.client = qdrant_client.QdrantClient(path=output)
-        self.client.set_model(EMBED_MODEL)
-        self.client.set_sparse_model(SPARSE_MODEL)
-        # optimizations to reduce ram
-        self.client.create_collection(
-            collection_name=COLLECTION_NAME,
-            vectors_config=self.client.get_fastembed_vector_params(on_disk=True),
-            sparse_vectors_config=self.client.get_fastembed_sparse_vector_params(on_disk=True),
-            quantization_config=models.ScalarQuantization(
-                scalar=models.ScalarQuantizationConfig(
-                    type=models.ScalarType.INT8,
-                    always_ram=True,
+        if self.format == "qdrant":
+            self.client = qdrant_client.QdrantClient(path=self.output)
+            self.client.set_model(EMBED_MODEL)
+            self.client.set_sparse_model(SPARSE_MODEL)
+            # optimizations to reduce ram
+            self.client.create_collection(
+                collection_name=COLLECTION_NAME,
+                vectors_config=self.client.get_fastembed_vector_params(on_disk=True),
+                sparse_vectors_config=self.client.get_fastembed_sparse_vector_params(on_disk=True),
+                quantization_config=models.ScalarQuantization(
+                    scalar=models.ScalarQuantizationConfig(
+                        type=models.ScalarType.INT8,
+                        always_ram=True,
+                    ),
                 ),
-            ),
-        )
+            )
 
     def add(self, file_path):
         if os.path.isdir(file_path):
             self.walk(file_path)  # Walk directory and process all files
         else:
-            self.targets.append(file_path)  # Process the single file
+            self.sources.append(file_path)  # Process the single file
 
-    def convert(self):
-        result = self.doc_converter.convert_all(self.targets)
+    def convert_qdrant(self, results):
         documents, ids = [], []
         chunker = HybridChunker(tokenizer=EMBED_MODEL, overlap=100, merge_peers=True)
-        for file in result:
-            chunk_iter = chunker.chunk(dl_doc=file.document)
+        for result in results:
+            chunk_iter = chunker.chunk(dl_doc=result.document)
             for chunk in chunk_iter:
                 # Extract the enriched text from the chunk
                 doc_text = chunker.contextualize(chunk=chunk)
@@ -72,6 +75,33 @@ class Converter:
                 ids.append(doc_id)
         return self.client.add(COLLECTION_NAME, documents=documents, ids=ids, batch_size=1)
 
+    def convert(self):
+        results = self.doc_converter.convert_all(self.sources)
+        if self.format == "qdrant":
+            return self.convert_qdrant(results)
+        if self.format == "markdown":
+            # Export the converted document to Markdown
+            return self.convert_markdown(results)
+        if self.format == "json":
+            # Export the converted document to JSON
+            return self.convert_json(results)
+
+    def convert_markdown(self, results):
+        ctr = 0
+        # Process the conversion results
+        for ctr, result in enumerate(results):
+            dirname = self.output + os.path.dirname(self.sources[ctr])
+            os.makedirs(dirname, exist_ok=True)
+            document = result.document
+            document.save_as_markdown(os.path.join(dirname, f"{document.name}.md"))
+
+    def convert_json(self, results):
+        for ctr, result in enumerate(results):
+            dirname = self.output + os.path.dirname(self.sources[ctr])
+            os.makedirs(dirname, exist_ok=True)
+            document = result.document
+            document.save_as_json(os.path.join(dirname, f"{document.name}.json"))
+
     def walk(self, path):
         for root, dirs, files in os.walk(path, topdown=True):
             if len(files) == 0:
@@ -79,7 +109,7 @@ class Converter:
             for f in files:
                 file = os.path.join(root, f)
                 if os.path.isfile(file):
-                    self.targets.append(file)
+                    self.sources.append(file)
 
     def generate_hash(self, document: str) -> str:
         """Generate a unique hash for a document."""
@@ -103,8 +133,14 @@ parser = argparse.ArgumentParser(
     description="process source files into RAG vector database",
 )
 
-parser.add_argument("target", nargs="?", help="Target database")
-parser.add_argument("source", nargs="*", help="Source files")
+parser.add_argument("output", nargs="?", help="Output directory")
+parser.add_argument("sources", nargs="*", help="Source files")
+parser.add_argument(
+    "--format",
+    default="qdrant",
+    help="Output format for RAG Data",
+    choices=["qdrant", "json", "markdown"],
+)
 parser.add_argument(
     "--ocr",
     action='store_true',
@@ -123,16 +159,16 @@ def eprint(e, exit_code):
 
 try:
     args = parser.parse_args()
-    if args.target == "load":
+    if args.output == "load":
         load()
     else:
-        converter = Converter(args.target, args.source, args.ocr)
+        converter = Converter(args)
         converter.convert()
 except docling.exceptions.ConversionError as e:
     eprint(e, 1)
 except FileNotFoundError as e:
-    eprint(e, 1)
+    eprint(e, errno.ENOENT)
 except ValueError as e:
-    eprint(e, 1)
+    eprint(e, errno.EINVAL)
 except KeyboardInterrupt:
     pass

container_build.sh

@@ -62,7 +62,7 @@ add_rag() {
     tag=$tag-rag
     containerfile="container-images/common/Containerfile.rag"
     GPU=cpu
-    case $2 in
+    case "${2##*/}" in
 	cuda)
 	    GPU=cuda
 	    ;;

docs/ramalama-bench.1.md

@@ -48,6 +48,33 @@ for a value and set the variable only if it is set on the host.
 #### **--help**, **-h**
 show this help message and exit
 
+#### **--image**=IMAGE
+OCI container image to run with specified AI model. RamaLama defaults to using
+images based on the accelerator it discovers. For example:
+`quay.io/ramalama/ramalama`. See the table below for all default images.
+The default image tag is based on the minor version of the RamaLama package.
+Version 0.11.0 of RamaLama pulls an image with a `:0.11` tag from the quay.io/ramalama OCI repository. The --image option overrides this default.
+
+The default can be overridden in the ramalama.conf file or via the
+RAMALAMA_IMAGE environment variable. `export RAMALAMA_IMAGE=quay.io/ramalama/aiimage:1.2` tells
+RamaLama to use the `quay.io/ramalama/aiimage:1.2` image.
+
+Accelerated images:
+
+| Accelerator             | Image                      |
+| ------------------------| -------------------------- |
+|  CPU, Apple             | quay.io/ramalama/ramalama  |
+|  HIP_VISIBLE_DEVICES    | quay.io/ramalama/rocm      |
+|  CUDA_VISIBLE_DEVICES   | quay.io/ramalama/cuda      |
+|  ASAHI_VISIBLE_DEVICES  | quay.io/ramalama/asahi     |
+|  INTEL_VISIBLE_DEVICES  | quay.io/ramalama/intel-gpu |
+|  ASCEND_VISIBLE_DEVICES | quay.io/ramalama/cann      |
+|  MUSA_VISIBLE_DEVICES   | quay.io/ramalama/musa      |
+
+#### **--keep-groups**
+pass --group-add keep-groups to podman (default: False)
+If GPU device on host system is accessible to user via group access, this option leaks the groups into the container.
+
 #### **--name**, **-n**
 name of the container to run the Model in
 
@@ -95,6 +122,9 @@ not have more privileges than the user that launched them.
 #### **--seed**=
 Specify seed rather than using random seed model interaction
 
+#### **--selinux**=*true*
+Enable SELinux container separation
+
 #### **--temp**="0.8"
 Temperature of the response from the AI Model
 llama.cpp explains this as:
@@ -118,7 +148,7 @@ Benchmark specified AI Model.
 ## EXAMPLES
 
 ```
-ramalama bench granite-moe3
+ramalama bench granite3-moe
 ```
 
 ## SEE ALSO

docs/ramalama-cuda.7.md

@@ -137,6 +137,19 @@ ramalama run granite
 
 This is particularly useful in multi-GPU systems where you want to dedicate specific GPUs to different workloads.
 
+If the `CUDA_VISIBLE_DEVICES` environment variable is set to an empty string, RamaLama will default to using the CPU.
+
+```bash
+export CUDA_VISIBLE_DEVICES=""  # Defaults to CPU
+ramalama run granite
+```
+
+To revert to using all available GPUs, unset the environment variable:
+
+```bash
+unset CUDA_VISIBLE_DEVICES
+```
+
 ## Troubleshooting
 
 ### CUDA Updates

docs/ramalama-perplexity.1.md

@@ -53,6 +53,33 @@ for a value and set the variable only if it is set on the host.
 #### **--help**, **-h**
 show this help message and exit
 
+#### **--image**=IMAGE
+OCI container image to run with specified AI model. RamaLama defaults to using
+images based on the accelerator it discovers. For example:
+`quay.io/ramalama/ramalama`. See the table below for all default images.
+The default image tag is based on the minor version of the RamaLama package.
+Version 0.11.0 of RamaLama pulls an image with a `:0.11` tag from the quay.io/ramalama OCI repository. The --image option overrides this default.
+
+The default can be overridden in the ramalama.conf file or via the
+RAMALAMA_IMAGE environment variable. `export RAMALAMA_IMAGE=quay.io/ramalama/aiimage:1.2` tells
+RamaLama to use the `quay.io/ramalama/aiimage:1.2` image.
+
+Accelerated images:
+
+| Accelerator             | Image                      |
+| ------------------------| -------------------------- |
+|  CPU, Apple             | quay.io/ramalama/ramalama  |
+|  HIP_VISIBLE_DEVICES    | quay.io/ramalama/rocm      |
+|  CUDA_VISIBLE_DEVICES   | quay.io/ramalama/cuda      |
+|  ASAHI_VISIBLE_DEVICES  | quay.io/ramalama/asahi     |
+|  INTEL_VISIBLE_DEVICES  | quay.io/ramalama/intel-gpu |
+|  ASCEND_VISIBLE_DEVICES | quay.io/ramalama/cann      |
+|  MUSA_VISIBLE_DEVICES   | quay.io/ramalama/musa      |
+
+#### **--keep-groups**
+pass --group-add keep-groups to podman (default: False)
+If GPU device on host system is accessible to user via group access, this option leaks the groups into the container.
+
 #### **--name**, **-n**
 name of the container to run the Model in
 
@@ -103,6 +130,9 @@ Add *args* to the runtime (llama.cpp or vllm) invocation.
 #### **--seed**=
 Specify seed rather than using random seed model interaction
 
+#### **--selinux**=*true*
+Enable SELinux container separation
+
 #### **--temp**="0.8"
 Temperature of the response from the AI Model
 llama.cpp explains this as:
@@ -126,7 +156,7 @@ Calculate the perplexity of an AI Model. Perplexity measures how well the model
 ## EXAMPLES
 
 ```
-ramalama perplexity granite-moe3
+ramalama perplexity granite3-moe
 ```
 
 ## SEE ALSO

docs/ramalama-rag.1.md

@@ -19,7 +19,7 @@ positional arguments:
 	    AsciiDoc & Markdown formatted files to be processed.
 	    Can be specified multiple times.
 
-  *IMAGE*   OCI Image name to contain processed rag data
+  *PATH|IMAGE*   Path or OCI Image name to contain processed rag data
 
 ## OPTIONS
 
@@ -32,9 +32,45 @@ process to be launched inside of the container. If an environment variable is
 specified without a value, the container engine checks the host environment
 for a value and set the variable only if it is set on the host.
 
+#### **--format**=*json* |  *markdown* | *qdrant* |
+Convert documents into the following formats
+
+| Type    | Description                                          |
+| ------- | ---------------------------------------------------- |
+| json    | JavaScript Object Notation. lightweight format for exchanging data |
+| markdown| Lightweight markup language using plain text editing |
+| qdrant  | Retrieval-Augmented Generation (RAG) Vector database |
+
 #### **--help**, **-h**
 Print usage message
 
+#### **--image**=IMAGE
+OCI container image to run with specified AI model. RamaLama defaults to using
+images based on the accelerator it discovers. For example:
+`quay.io/ramalama/ramalama-rag`. See the table below for all default images.
+The default image tag is based on the minor version of the RamaLama package.
+Version 0.11.0 of RamaLama pulls an image with a `:0.11` tag from the quay.io/ramalama OCI repository. The --image option overrides this default.
+
+The default can be overridden in the ramalama.conf file or via the
+RAMALAMA_IMAGE environment variable. `export RAMALAMA_IMAGE=quay.io/ramalama/aiimage:1.2` tells
+RamaLama to use the `quay.io/ramalama/aiimage:1.2` image.
+
+Accelerated images:
+
+| Accelerator             | Image                          |
+| ------------------------| ------------------------------ |
+|  CPU, Apple             | quay.io/ramalama/ramalama-rag  |
+|  HIP_VISIBLE_DEVICES    | quay.io/ramalama/rocm-rag      |
+|  CUDA_VISIBLE_DEVICES   | quay.io/ramalama/cuda-rag      |
+|  ASAHI_VISIBLE_DEVICES  | quay.io/ramalama/asahi-rag     |
+|  INTEL_VISIBLE_DEVICES  | quay.io/ramalama/intel-gpu-rag |
+|  ASCEND_VISIBLE_DEVICES | quay.io/ramalama/cann-rag      |
+|  MUSA_VISIBLE_DEVICES   | quay.io/ramalama/musa-rag      |
+
+#### **--keep-groups**
+pass --group-add keep-groups to podman (default: False)
+If GPU device on host system is accessible to user via group access, this option leaks the groups into the container.
+
 #### **--network**=*none*
 sets the configuration for network namespaces when handling RUN instructions
 
@@ -49,10 +85,13 @@ Pull image policy. The default is **missing**.
 - **never**: Never pull the image but use the one from the local containers storage. Throw an error when no image is found.
 - **newer**: Pull if the image on the registry is newer than the one in the local containers storage. An image is considered to be newer when the digests are different. Comparing the time stamps is prone to errors. Pull errors are suppressed if a local image was found.
 
+#### **--selinux**=*true*
+Enable SELinux container separation
+
 ## EXAMPLES
 
 ```
-./bin/ramalama rag ./README.md https://github.com/containers/podman/blob/main/README.md quay.io/rhatdan/myrag
+$ ramalama rag ./README.md https://github.com/containers/podman/blob/main/README.md quay.io/rhatdan/myrag
 100% |███████████████████████████████████████████████████████|  114.00 KB/    0.00 B 922.89 KB/s   59m 59s
 Building quay.io/ramalama/myrag...
 adding vectordb...
@@ -60,7 +99,17 @@ c857ebc65c641084b34e39b740fdb6a2d9d2d97be320e6aa9439ed0ab8780fe0
 ```
 
 ```
-ramalama rag --ocr README.md https://mysight.edu/document quay.io/rhatdan/myrag
+$ ramalama rag --ocr README.md https://mysight.edu/document quay.io/rhatdan/myrag
+```
+
+```
+$ ramalama rag --format markdown /tmp/internet.pdf /tmp/output
+$ ls /tmp/output/docs/tmp/
+/tmp/output/docs/tmp/internet.md
+$ ramalama rag --format json /tmp/internet.pdf /tmp/output
+$ ls /tmp/output/docs/tmp/
+/tmp/output/docs/tmp/internet.md
+/tmp/output/docs/tmp/internet.json
 ```
 
 ## SEE ALSO

docs/ramalama-run.1.md

@@ -61,6 +61,33 @@ for a value and set the variable only if it is set on the host.
 #### **--help**, **-h**
 Show this help message and exit
 
+#### **--image**=IMAGE
+OCI container image to run with specified AI model. RamaLama defaults to using
+images based on the accelerator it discovers. For example:
+`quay.io/ramalama/ramalama`. See the table below for all default images.
+The default image tag is based on the minor version of the RamaLama package.
+Version 0.11.0 of RamaLama pulls an image with a `:0.11` tag from the quay.io/ramalama OCI repository. The --image option overrides this default.
+
+The default can be overridden in the ramalama.conf file or via the
+RAMALAMA_IMAGE environment variable. `export RAMALAMA_IMAGE=quay.io/ramalama/aiimage:1.2` tells
+RamaLama to use the `quay.io/ramalama/aiimage:1.2` image.
+
+Accelerated images:
+
+| Accelerator             | Image                      |
+| ------------------------| -------------------------- |
+|  CPU, Apple             | quay.io/ramalama/ramalama  |
+|  HIP_VISIBLE_DEVICES    | quay.io/ramalama/rocm      |
+|  CUDA_VISIBLE_DEVICES   | quay.io/ramalama/cuda      |
+|  ASAHI_VISIBLE_DEVICES  | quay.io/ramalama/asahi     |
+|  INTEL_VISIBLE_DEVICES  | quay.io/ramalama/intel-gpu |
+|  ASCEND_VISIBLE_DEVICES | quay.io/ramalama/cann      |
+|  MUSA_VISIBLE_DEVICES   | quay.io/ramalama/musa      |
+
+#### **--keep-groups**
+pass --group-add keep-groups to podman (default: False)
+If GPU device on host system is accessible to user via group access, this option leaks the groups into the container.
+
 #### **--keepalive**
 duration to keep a model loaded (e.g. 5m)
 
@@ -121,6 +148,9 @@ Add *args* to the runtime (llama.cpp or vllm) invocation.
 #### **--seed**=
 Specify seed rather than using random seed model interaction
 
+#### **--selinux**=*true*
+Enable SELinux container separation
+
 #### **--temp**="0.8"
 Temperature of the response from the AI Model
 llama.cpp explains this as:

docs/ramalama-serve.1.md

@@ -93,6 +93,33 @@ show this help message and exit
 #### **--host**="0.0.0.0"
 IP address for llama.cpp to listen on.
 
+#### **--image**=IMAGE
+OCI container image to run with specified AI model. RamaLama defaults to using
+images based on the accelerator it discovers. For example:
+`quay.io/ramalama/ramalama`. See the table above for all default images.
+The default image tag is based on the minor version of the RamaLama package.
+Version 0.11.0 of RamaLama pulls an image with a `:0.11` tag from the quay.io/ramalama OCI repository. The --image option overrides this default.
+
+The default can be overridden in the ramalama.conf file or via the
+RAMALAMA_IMAGE environment variable. `export RAMALAMA_IMAGE=quay.io/ramalama/aiimage:1.2` tells
+RamaLama to use the `quay.io/ramalama/aiimage:1.2` image.
+
+Accelerated images:
+
+| Accelerator             | Image                      |
+| ------------------------| -------------------------- |
+|  CPU, Apple             | quay.io/ramalama/ramalama  |
+|  HIP_VISIBLE_DEVICES    | quay.io/ramalama/rocm      |
+|  CUDA_VISIBLE_DEVICES   | quay.io/ramalama/cuda      |
+|  ASAHI_VISIBLE_DEVICES  | quay.io/ramalama/asahi     |
+|  INTEL_VISIBLE_DEVICES  | quay.io/ramalama/intel-gpu |
+|  ASCEND_VISIBLE_DEVICES | quay.io/ramalama/cann      |
+|  MUSA_VISIBLE_DEVICES   | quay.io/ramalama/musa      |
+
+#### **--keep-groups**
+pass --group-add keep-groups to podman (default: False)
+If GPU device on host system is accessible to user via group access, this option leaks the groups into the container.
+
 #### **--model-draft**
 
 A draft model is a smaller, faster model that helps accelerate the decoding
@@ -164,6 +191,9 @@ Add *args* to the runtime (llama.cpp or vllm) invocation.
 #### **--seed**=
 Specify seed rather than using random seed model interaction
 
+#### **--selinux**=*true*
+Enable SELinux container separation
+
 #### **--temp**="0.8"
 Temperature of the response from the AI Model.
 llama.cpp explains this as: