Merge branch 'opea-project:main' into main

.github/license_template.txt

@@ -1,2 +1,2 @@
-Copyright (C) 2024 Intel Corporation
+Copyright (C) 2025 Intel Corporation
 SPDX-License-Identifier: Apache-2.0

.github/workflows/pr-chart-e2e.yaml

@@ -34,6 +34,10 @@ jobs:
       - name: Get test matrix
         id: get-test-matrix
         run: |
+          if grep -r '^version: ' `find helm-charts/ -name "Chart.yaml"` | grep -v "0-latest"; then
+            echo "Version check failed, main branch should always use version 0-latest"
+            exit 1
+          fi
           base_commit=${{ github.event.pull_request.base.sha }}
           merged_commit=$(git log -1 --format='%H')
           e2e_charts=$(git diff --name-only ${base_commit} ${merged_commit} | \

.github/workflows/releasedevhelm.yaml

@@ -0,0 +1,66 @@
+# Copyright (C) 2025 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+name: Update the helm charts 0-latest
+
+on:
+  pull_request:
+    branches:
+      - main
+    types: [closed]
+    paths:
+      - helm-charts/**
+env:
+  CHARTS_DIR: "helm-charts"
+
+jobs:
+  run-on-merge:
+    if: github.event.pull_request.merged == true
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout Code
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          ref: ${{ github.ref }}
+
+      - name: Get the modified charts list
+        run: |
+          echo "This job runs after a PR is merged!"
+          if grep -r '^version: ' `find helm-charts/ -name "Chart.yaml"` | grep -v "0-latest"; then
+            echo "Version check failed, only update 0-latest"
+            exit 1
+          fi
+          base_commit=${{ github.event.pull_request.base.sha }}
+          merged_commit=$(git log -1 --format='%H')
+          # Update components
+          common_charts=$(git diff --name-only ${base_commit} ${merged_commit} | \
+          grep "^$CHARTS_DIR/common" | \
+          grep -vE 'README.md|*.sh' | \
+          cut -d'/' -f3 | sort -u )
+          echo "Charts to be updated: $common_charts"
+          echo "${{ secrets.ACTION_TOKEN }}" | helm registry login ghcr.io -u opea --password-stdin
+          pushd $CHARTS_DIR/common
+          for chart in ${common_charts}; do
+            echo "Updating $chart"
+            helm dependency update ${chart}
+            helm package $chart
+            helm push ${chart}-0-latest.tgz oci://ghcr.io/opea-project/charts
+          done
+          popd
+          # Update Examples
+          e2e_charts=$(git diff --name-only ${base_commit} ${merged_commit} | \
+          grep -v "common" | \
+          grep -vE 'README.md|*.sh' | \
+          cut -d'/' -f2 | sort -u )
+          echo "Charts to be updated: $e2e_charts"
+          echo "${{ secrets.ACTION_TOKEN }}" | helm registry login ghcr.io -u opea --password-stdin
+          pushd $CHARTS_DIR
+          for chart in ${e2e_charts}; do
+            echo "Updating $chart"
+            helm dependency update ${chart}
+            helm package $chart
+            helm push ${chart}-0-latest.tgz oci://ghcr.io/opea-project/charts
+          done
+          popd

helm-charts/audioqna/Chart.yaml

@@ -6,27 +6,21 @@ name: audioqna
 description: The Helm chart to deploy AudioQnA
 type: application
 dependencies:
-  - name: asr
-    version: 0-latest
-    repository: "file://../common/asr"
   - name: whisper
     version: 0-latest
     repository: "file://../common/whisper"
-  - name: tts
-    version: 0-latest
-    repository: "file://../common/tts"
   - name: speecht5
     version: 0-latest
     repository: "file://../common/speecht5"
   - name: tgi
     version: 0-latest
     repository: "file://../common/tgi"
-  - name: llm-uservice
-    version: 0-latest
-    repository: "file://../common/llm-uservice"
   - name: ui
     version: 0-latest
     repository: "file://../common/ui"
     alias: audioqna-ui
+  - name: nginx
+    version: 0-latest
+    repository: "file://../common/nginx"
 version: 0-latest
 appVersion: "1.1"

helm-charts/audioqna/gaudi-values.yaml

@@ -12,6 +12,12 @@ tgi:
   MAX_INPUT_LENGTH: "1024"
   MAX_TOTAL_TOKENS: "2048"
   CUDA_GRAPHS: ""
+  HF_HUB_DISABLE_PROGRESS_BARS: 1
+  HF_HUB_ENABLE_HF_TRANSFER: 0
+  ENABLE_HPU_GRAPH: true
+  LIMIT_HPU_GRAPH: true
+  USE_FLASH_ATTENTION: true
+  FLASH_ATTENTION_RECOMPUTE: true
   livenessProbe:
     initialDelaySeconds: 5
     periodSeconds: 5

helm-charts/audioqna/templates/deployment.yaml

@@ -34,18 +34,18 @@ spec:
       containers:
         - name: {{ .Release.Name }}
           env:
-            - name: LLM_SERVICE_HOST_IP
-              value: {{ include "llm-uservice.fullname" (index .Subcharts "llm-uservice") }}
-            - name: LLM_SERVICE_PORT
-              value: {{ index .Values "llm-uservice" "service" "port" | quote }}
-            - name: ASR_SERVICE_HOST_IP
-              value: {{ include "asr.fullname" (index .Subcharts "asr") }}
-            - name: ASR_SERVICE_PORT
-              value: {{ index .Values "asr" "service" "port" | quote }}
-            - name: TTS_SERVICE_HOST_IP
-              value: {{ include "tts.fullname" (index .Subcharts "tts") }}
-            - name: TTS_SERVICE_PORT
-              value: {{ index .Values "tts" "service" "port" | quote }}
+            - name: LLM_SERVER_HOST_IP
+              value: {{ include "tgi.fullname" (index .Subcharts "tgi") }}
+            - name: LLM_SERVER_PORT
+              value: "80"
+            - name: WHISPER_SERVER_HOST_IP
+              value: {{ include "whisper.fullname" (index .Subcharts "whisper") }}
+            - name: WHISPER_SERVER_PORT
+              value: {{ index .Values "whisper" "service" "port" | quote }}
+            - name: SPEECHT5_SERVER_HOST_IP
+              value: {{ include "speecht5.fullname" (index .Subcharts "speecht5") }}
+            - name: SPEECHT5_SERVER_PORT
+              value: {{ index .Values "speecht5" "service" "port" | quote }}
           securityContext:
             {{- toYaml .Values.securityContext | nindent 12 }}
           image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"

helm-charts/audioqna/templates/nginx.yaml

@@ -0,0 +1,15 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ .Release.Name }}-nginx-config
+  labels:
+    {{- include "audioqna.labels" . | nindent 4 }}
+data:
+  FRONTEND_SERVICE_IP: {{ include "ui.fullname" (index .Subcharts "audioqna-ui") | quote }}
+  FRONTEND_SERVICE_PORT: {{ index .Values "audioqna-ui" "service" "port" | quote }}
+  BACKEND_SERVICE_NAME: audioqna
+  BACKEND_SERVICE_IP: {{ include "audioqna.fullname" . | quote }}
+  BACKEND_SERVICE_PORT: {{ .Values.service.port | quote }}

helm-charts/audioqna/values.yaml

@@ -50,10 +50,6 @@ service:
   type: ClusterIP
   port: 3008
 
-nginx:
-  service:
-    type: NodePort
-
 nodeSelector: {}
 
 tolerations: []
@@ -64,6 +60,10 @@ affinity: {}
 tgi:
   LLM_MODEL_ID: Intel/neural-chat-7b-v3-3
 
+nginx:
+  service:
+    type: NodePort
+
 audioqna-ui:
   image:
     repository: opea/audioqna-ui

helm-charts/chatqna/Chart.yaml

@@ -10,7 +10,7 @@ dependencies:
     version: 0-latest
     alias: tgi-guardrails
     repository: "file://../common/tgi"
-    condition: guardrails-usvc.enabled
+    condition: tgi-guardrails.enabled
   - name: guardrails-usvc
     version: 0-latest
     repository: "file://../common/guardrails-usvc"
@@ -23,15 +23,29 @@ dependencies:
     version: 0-latest
     repository: "file://../common/vllm"
     condition: vllm.enabled
+  - name: llm-uservice
+    version: 0-latest
+    repository: "file://../common/llm-uservice"
+    condition: llm-uservice.enabled
   - name: tei
     version: 0-latest
     repository: "file://../common/tei"
+  - name: embedding-usvc
+    version: 0-latest
+    repository: "file://../common/embedding-usvc"
+    condition: embedding-usvc.enabled
   - name: teirerank
     version: 0-latest
     repository: "file://../common/teirerank"
+    condition: teirerank.enabled
+  - name: reranking-usvc
+    version: 0-latest
+    repository: "file://../common/reranking-usvc"
+    condition: reranking-usvc.enabled
   - name: redis-vector-db
     version: 0-latest
     repository: "file://../common/redis-vector-db"
+    condition: redis-vector-db.enabled
   - name: retriever-usvc
     version: 0-latest
     repository: "file://../common/retriever-usvc"
@@ -42,5 +56,8 @@ dependencies:
     alias: chatqna-ui
     version: 0-latest
     repository: "file://../common/ui"
+  - name: nginx
+    version: 0-latest
+    repository: "file://../common/nginx"
 version: 0-latest
 appVersion: "v1.0"

helm-charts/chatqna/guardrails-gaudi-values.yaml

@@ -77,6 +77,7 @@ tgi:
     failureThreshold: 120
 
 tgi-guardrails:
+  enabled: true
   accelDevice: "gaudi"
   LLM_MODEL_ID: "meta-llama/Meta-Llama-Guard-2-8B"
   image:

helm-charts/chatqna/guardrails-values.yaml

@@ -11,4 +11,5 @@ guardrails-usvc:
   # SAFETY_GUARD_ENDPOINT: "http://{{ .Release.Name }}-tgi-guardrails"
   SAFETY_GUARD_MODEL_ID: "meta-llama/Meta-Llama-Guard-2-8B"
 tgi-guardrails:
+  enabled: true
   LLM_MODEL_ID: "meta-llama/Meta-Llama-Guard-2-8B"

helm-charts/chatqna/norerank-values.yaml

@@ -0,0 +1,11 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+# Accelerate inferencing in heaviest components to improve performance
+# by overriding their subchart values
+
+image:
+  repository: opea/chatqna-without-rerank
+
+teirerank:
+  enabled: false

helm-charts/chatqna/templates/deployment.yaml

@@ -64,6 +64,18 @@ spec:
               value: {{ .Release.Name }}-guardrails-usvc
             - name: GUARDRAIL_SERVICE_PORT
               value: "9090"
+            - name: LLM_SERVICE_HOST_IP
+              value: {{ .Release.Name }}-llm-uservice
+            - name: LLM_SERVICE_PORT
+              value: "9000"
+            - name: RERANK_SERVICE_HOST_IP
+              value: {{ .Release.Name }}-reranking-usvc
+            - name: RERANK_SERVICE_PORT
+              value: "8000"
+            - name: EMBEDDING_SERVICE_HOST_IP
+              value: {{ .Release.Name }}-embedding-usvc
+            - name: EMBEDDING_SERVICE_PORT
+              value: "6000"
             {{- if .Values.LOGFLAG }}
             - name: LOGFLAG
               value: {{ .Values.LOGFLAG | quote }}