From 3ec5fbae15ac11747c78ffb6c2fe8dae19cc3915 Mon Sep 17 00:00:00 2001
From: Administrator <admin@example.com>
Date: Wed, 17 Dec 2025 17:23:55 +0800
Subject: [PATCH 1/9] Support PolyLingua on Gaudi Platform

---
 PolyLingua/.env.example                       |   3 +-
 PolyLingua/README.md                          |   2 +-
 .../intel/hpu/gaudi/compose.yaml              | 106 +++++
 PolyLingua/set_env.sh                         |   3 +
 PolyLingua/tests/test_compose_on_gaudi.sh     | 361 ++++++++++++++++++
 5 files changed, 473 insertions(+), 2 deletions(-)
 create mode 100644 PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
 create mode 100755 PolyLingua/tests/test_compose_on_gaudi.sh

diff --git a/PolyLingua/.env.example b/PolyLingua/.env.example
index a48a3de7cd..0c72cc3459 100644
--- a/PolyLingua/.env.example
+++ b/PolyLingua/.env.example
@@ -18,7 +18,8 @@ HF_TOKEN=hf_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
 # LLM model ID from HuggingFace
 # Default model supports multilingual translation
 LLM_MODEL_ID=swiss-ai/Apertus-8B-Instruct-2509
-
+# Gaudi CARDS
+NUM_CARDS=1
 # Directory to cache downloaded models
 # Models can be large (several GB), ensure sufficient disk space
 MODEL_CACHE=./data
diff --git a/PolyLingua/README.md b/PolyLingua/README.md
index 26bff952f9..5b5cbba5b4 100644
--- a/PolyLingua/README.md
+++ b/PolyLingua/README.md
@@ -111,7 +111,7 @@ The service works with any HuggingFace text generation model. Recommended models
 
 - **swiss-ai/Apertus-8B-Instruct-2509** - Multilingual translation (default)
 - **haoranxu/ALMA-7B** - Specialized translation model
-
+- **Qwen/Qwen2.5-7B-Instruct88 - Other Common model (Gaudi default)
 ## 🛠️ Development
 
 ### Project Structure
diff --git a/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
new file mode 100644
index 0000000000..5878d36fb8
--- /dev/null
+++ b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
@@ -0,0 +1,106 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+services:
+  vllm-service:
+    image: opea/vllm-gaudi:1.4
+    container_name: vllm-gaudi-server
+    ports:
+      - "8028:80"
+    volumes:
+      - "${MODEL_CACHE:-./data}:/root/.cache/huggingface/hub"
+    shm_size: 1g
+    environment:
+      no_proxy: ${no_proxy}
+      http_proxy: ${http_proxy}
+      https_proxy: ${https_proxy}
+      HF_TOKEN: ${HF_TOKEN}
+      HABANA_VISIBLE_DEVICES: all
+      OMPI_MCA_btl_vader_single_copy_mechanism: none
+      VLLM_SKIP_WARMUP: ${VLLM_SKIP_WARMUP:-true}
+      NUM_CARDS: ${NUM_CARDS:-1}
+      VLLM_TORCH_PROFILER_DIR: "/mnt"
+    healthcheck:
+      test: ["CMD-SHELL", "curl -f http://localhost:80/health || exit 1"]
+      interval: 10s
+      timeout: 10s
+      retries: 100
+    runtime: habana
+    cap_add:
+      - SYS_NICE
+    ipc: host
+    command: --model ${LLM_MODEL_ID} --tensor-parallel-size ${NUM_CARDS} --host 0.0.0.0 --port 80 --block-size 128 --max-num-seqs 256
+  llm:
+    image: ${REGISTRY:-opea}/llm-textgen:${TAG:-latest}
+    container_name: llm-textgen-server
+    depends_on:
+      vllm-service:
+        condition: service_healthy
+    ports:
+      - "9000:9000"
+    ipc: host
+    environment:
+      no_proxy: ${no_proxy}
+      http_proxy: ${http_proxy}
+      https_proxy: ${https_proxy}
+      LLM_ENDPOINT: ${VLLM_ENDPOINT}
+      LLM_MODEL_ID: ${LLM_MODEL_ID}
+      HF_TOKEN: ${HF_TOKEN}
+      HF_HUB_DISABLE_PROGRESS_BARS: 1
+      HF_HUB_ENABLE_HF_TRANSFER: 0
+    restart: unless-stopped
+
+  polylingua-xeon-backend-server:
+    image: ${REGISTRY:-opea}/polylingua:${TAG:-latest}
+    container_name: polylingua-xeon-backend-server
+    depends_on:
+      - vllm-service
+      - llm
+    ports:
+      - "8888:8888"
+    environment:
+      - no_proxy=${no_proxy}
+      - https_proxy=${https_proxy}
+      - http_proxy=${http_proxy}
+      - MEGA_SERVICE_HOST_IP=${MEGA_SERVICE_HOST_IP}
+      - MEGA_SERVICE_PORT=8888
+      - LLM_SERVICE_HOST_IP=${LLM_SERVICE_HOST_IP}
+      - LLM_SERVICE_PORT=${LLM_SERVICE_PORT}
+    ipc: host
+    restart: always
+
+  polylingua-ui-server:
+    image: ${REGISTRY:-opea}/polylingua-ui:${TAG:-latest}
+    container_name: polylingua-ui-server
+    depends_on:
+      - polylingua-xeon-backend-server
+    ports:
+      - "5173:5173"
+    environment:
+      - no_proxy=${no_proxy}
+      - https_proxy=${https_proxy}
+      - http_proxy=${http_proxy}
+      - BACKEND_SERVICE_ENDPOINT=${BACKEND_SERVICE_ENDPOINT}
+    ipc: host
+    restart: always
+
+  polylingua-nginx-server:
+    image: nginx:alpine
+    container_name: polylingua-nginx-server
+    depends_on:
+      - polylingua-xeon-backend-server
+      - polylingua-ui-server
+    ports:
+      - "${NGINX_PORT:-80}:80"
+    volumes:
+      - ../../../../deploy/nginx.conf:/etc/nginx/nginx.conf:ro
+    environment:
+      - no_proxy=${no_proxy}
+      - https_proxy=${https_proxy}
+      - http_proxy=${http_proxy}
+    ipc: host
+    restart: always
+
+networks:
+  default:
+    driver: bridge
diff --git a/PolyLingua/set_env.sh b/PolyLingua/set_env.sh
index 3a56740ed5..56b6cef019 100755
--- a/PolyLingua/set_env.sh
+++ b/PolyLingua/set_env.sh
@@ -81,6 +81,9 @@ echo "export BACKEND_SERVICE_IP=\"${host_ip}\"" >> .env
 export BACKEND_SERVICE_PORT="8888"
 echo "export BACKEND_SERVICE_PORT=\"8888\"" >> .env
 
+export NUM_CARDS="1"
+echo "export NUM_CARDS=\"1\"" >> .env
+
 # Docker Configuration
 echo ""
 echo "--- Docker Configuration ---"
diff --git a/PolyLingua/tests/test_compose_on_gaudi.sh b/PolyLingua/tests/test_compose_on_gaudi.sh
new file mode 100755
index 0000000000..976d1575f1
--- /dev/null
+++ b/PolyLingua/tests/test_compose_on_gaudi.sh
@@ -0,0 +1,361 @@
+#!/bin/bash
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+set -xe
+IMAGE_REPO=${IMAGE_REPO:-"opea"}
+IMAGE_TAG=${IMAGE_TAG:-"latest"}
+echo "REGISTRY=IMAGE_REPO=${IMAGE_REPO}"
+echo "TAG=IMAGE_TAG=${IMAGE_TAG}"
+export REGISTRY=${IMAGE_REPO}
+export TAG=${IMAGE_TAG}
+export MODEL_CACHE=${model_cache:-"./data"}
+
+# Get the directory where this script is located
+SCRIPT_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# PolyLingua root is one level up from tests directory
+WORKPATH=$(dirname "$SCRIPT_DIR")
+LOG_PATH="$WORKPATH/tests"
+ip_address=$(hostname -I | awk '{print $1}')
+
+echo "Script directory: $SCRIPT_DIR"
+echo "Working directory: $WORKPATH"
+
+function build_docker_images() {
+    opea_branch=${opea_branch:-"main"}
+    cd $WORKPATH/docker_image_build
+
+    # Clone GenAIComps
+    git clone --depth 1 --branch ${opea_branch} https://github.com/opea-project/GenAIComps.git
+    pushd GenAIComps
+    echo "GenAIComps test commit is $(git rev-parse HEAD)"
+    docker build --no-cache -t ${REGISTRY}/comps-base:${TAG} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile .
+    popd && sleep 1s
+
+    # Build all images using build.yaml
+    echo "Building PolyLingua images with --no-cache, check docker_image_build.log for details..."
+    service_list="polylingua polylingua-ui llm-textgen"
+    docker compose -f build.yaml build ${service_list} --no-cache > ${LOG_PATH}/docker_image_build.log 2>&1
+
+    echo "Image build completed"
+    docker images | grep -E "polylingua|llm-textgen"
+    sleep 1s
+}
+
+function start_services() {
+    cd $WORKPATH/docker_compose/intel/hpu/gaudi/
+    export host_ip=${ip_address}
+    export no_proxy="localhost,127.0.0.1,$ip_address"
+
+    # Load environment variables
+    if [ ! -f .env ]; then
+        echo "Creating .env file..."
+        export HF_TOKEN=${HF_TOKEN}
+        export LLM_MODEL_ID="Qwen/Qwen2.5-7B-Instruct"
+        export VLLM_ENDPOINT="http://${host_ip}:8028"
+        export LLM_SERVICE_HOST_IP=${host_ip}
+        export LLM_SERVICE_PORT=9000
+        export MEGA_SERVICE_HOST_IP=${host_ip}
+        export MEGA_SERVICE_PORT=8888
+        export BACKEND_SERVICE_ENDPOINT="http://${host_ip}:8888"
+        export BACKEND_SERVICE_NAME="polylingua"
+        export BACKEND_SERVICE_IP=${host_ip}
+        export BACKEND_SERVICE_PORT=8888
+        export FRONTEND_SERVICE_IP=${host_ip}
+        export FRONTEND_SERVICE_PORT=5173
+        export NGINX_PORT=80
+
+        cat > .env <<EOF
+HF_TOKEN=${HF_TOKEN}
+LLM_MODEL_ID=${LLM_MODEL_ID}
+MODEL_CACHE=${MODEL_CACHE}
+host_ip=${host_ip}
+VLLM_ENDPOINT=${VLLM_ENDPOINT}
+LLM_SERVICE_HOST_IP=${LLM_SERVICE_HOST_IP}
+LLM_SERVICE_PORT=${LLM_SERVICE_PORT}
+MEGA_SERVICE_HOST_IP=${MEGA_SERVICE_HOST_IP}
+MEGA_SERVICE_PORT=${MEGA_SERVICE_PORT}
+BACKEND_SERVICE_ENDPOINT=${BACKEND_SERVICE_ENDPOINT}
+BACKEND_SERVICE_NAME=${BACKEND_SERVICE_NAME}
+BACKEND_SERVICE_IP=${BACKEND_SERVICE_IP}
+BACKEND_SERVICE_PORT=${BACKEND_SERVICE_PORT}
+FRONTEND_SERVICE_IP=${FRONTEND_SERVICE_IP}
+FRONTEND_SERVICE_PORT=${FRONTEND_SERVICE_PORT}
+NGINX_PORT=${NGINX_PORT}
+REGISTRY=${REGISTRY}
+TAG=${TAG}
+NUM_CARDS=1
+EOF
+    fi
+
+    # Start Docker Containers
+    echo "Starting services with docker compose..."
+    docker compose -f compose.yaml up -d > ${LOG_PATH}/start_services_with_compose.log 2>&1
+
+    # Wait for vLLM service to be ready
+    echo "Waiting for vLLM service to initialize (this may take several minutes)..."
+    n=0
+    until [[ "$n" -ge 100 ]]; do
+        docker logs vllm-service > ${LOG_PATH}/vllm_service_start.log 2>&1
+        if grep -E "Uvicorn running|Application startup complete" ${LOG_PATH}/vllm_service_start.log; then
+            echo "vLLM service is ready!"
+            break
+        fi
+        if grep -q "error" ${LOG_PATH}/vllm_service_start.log; then
+            echo "Error detected in vLLM service startup"
+            cat ${LOG_PATH}/vllm_service_start.log
+            exit 1
+        fi
+        sleep 10s
+        n=$((n+1))
+    done
+
+    if [[ "$n" -ge 100 ]]; then
+        echo "Timeout waiting for vLLM service"
+        docker logs vllm-service
+        exit 1
+    fi
+
+    echo "Waiting additional 10s for all services to stabilize..."
+    sleep 10s
+}
+
+function validate_services() {
+    local URL="$1"
+    local EXPECTED_RESULT="$2"
+    local SERVICE_NAME="$3"
+    local DOCKER_NAME="$4"
+    local INPUT_DATA="$5"
+    local CONTENT_TYPE="${6:-application/json}"
+
+    echo "Testing $SERVICE_NAME at $URL"
+
+    if [[ "$CONTENT_TYPE" == "multipart/form-data" ]]; then
+        # Handle file upload
+        local HTTP_STATUS=$(eval curl -s -o /dev/null -w "%{http_code}" -X POST $INPUT_DATA "$URL")
+
+        if [ "$HTTP_STATUS" -eq 200 ]; then
+            echo "[ $SERVICE_NAME ] HTTP status is 200. Checking content..."
+            local CONTENT=$(eval curl -s -X POST $INPUT_DATA "$URL" | tee ${LOG_PATH}/${SERVICE_NAME}.log)
+
+            if echo "$CONTENT" | grep -q "$EXPECTED_RESULT"; then
+                echo "[ $SERVICE_NAME ] ✓ Content is as expected."
+            else
+                echo "[ $SERVICE_NAME ] ✗ Content does not match expected result"
+                echo "Expected: $EXPECTED_RESULT"
+                echo "Got: $CONTENT"
+                docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+                exit 1
+            fi
+        else
+            echo "[ $SERVICE_NAME ] ✗ HTTP status is $HTTP_STATUS (expected 200)"
+            docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+            exit 1
+        fi
+    else
+        # Handle JSON request
+        local HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" -X POST -d "$INPUT_DATA" -H "Content-Type: $CONTENT_TYPE" "$URL")
+
+        if [ "$HTTP_STATUS" -eq 200 ]; then
+            echo "[ $SERVICE_NAME ] HTTP status is 200. Checking content..."
+
+            local CONTENT=$(curl -s -X POST -d "$INPUT_DATA" -H "Content-Type: $CONTENT_TYPE" "$URL" | tee ${LOG_PATH}/${SERVICE_NAME}.log)
+
+            if echo "$CONTENT" | grep -q "$EXPECTED_RESULT"; then
+                echo "[ $SERVICE_NAME ] ✓ Content is as expected."
+            else
+                echo "[ $SERVICE_NAME ] ✗ Content does not match expected result"
+                echo "Expected: $EXPECTED_RESULT"
+                echo "Got: $CONTENT"
+                docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+                exit 1
+            fi
+        else
+            echo "[ $SERVICE_NAME ] ✗ HTTP status is $HTTP_STATUS (expected 200)"
+            docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+            exit 1
+        fi
+    fi
+
+    sleep 2s
+}
+
+function validate_microservices() {
+    echo "======================================"
+    echo "Validating Microservices"
+    echo "======================================"
+
+    # Test vLLM service health
+    echo "Testing vLLM service health..."
+    curl -s http://${ip_address}:8028/health || {
+        echo "vLLM health check failed"
+        exit 1
+    }
+    echo "✓ vLLM service health check passed"
+
+    # Test vLLM service chat completions
+    validate_services \
+        "http://${ip_address}:8028/v1/chat/completions" \
+        "content" \
+        "vllm" \
+        "vllm-service" \
+        '{"model": "Qwen/Qwen2.5-7B-Instruct", "messages": [{"role": "user", "content": "Translate Hello to Spanish"}], "max_tokens": 32}'
+
+    # Test LLM microservice
+    validate_services \
+        "http://${ip_address}:9000/v1/chat/completions" \
+        "data: " \
+        "llm" \
+        "llm-textgen-server" \
+        '{"query":"Translate Hello to Spanish", "max_tokens": 32}'
+}
+
+function validate_megaservice() {
+    echo "======================================"
+    echo "Validating Megaservice"
+    echo "======================================"
+
+    # Test 1: Basic text translation (English to Spanish)
+    echo "Test 1: Basic English to Spanish translation..."
+    validate_services \
+        "http://${ip_address}:8888/v1/translation" \
+        "choices" \
+        "mega-polylingua-basic" \
+        "polylingua-xeon-backend-server" \
+        '{"language_from": "English", "language_to": "Spanish", "source_language": "Hello, how are you today?"}'
+
+    # Test 2: Language auto-detection
+    echo "Test 2: Auto-detection test..."
+    validate_services \
+        "http://${ip_address}:8888/v1/translation" \
+        "choices" \
+        "mega-polylingua-auto" \
+        "polylingua-xeon-backend-server" \
+        '{"language_from": "auto", "language_to": "French", "source_language": "Hello world"}'
+
+    # Test 3: Different language pair (English to German)
+    echo "Test 3: English to German translation..."
+    validate_services \
+        "http://${ip_address}:8888/v1/translation" \
+        "choices" \
+        "mega-polylingua-german" \
+        "polylingua-xeon-backend-server" \
+        '{"language_from": "English", "language_to": "German", "source_language": "Good morning"}'
+}
+
+function validate_file_translation() {
+    echo "======================================"
+    echo "Validating File Upload Translation"
+    echo "======================================"
+
+    # Create test file
+    cd $WORKPATH/tests
+    mkdir -p test_data
+    echo "Hello, this is a test document for translation. It contains multiple sentences. We want to test if file upload works correctly." > test_data/sample.txt
+
+    # Test file upload translation
+    echo "Testing file upload translation..."
+    validate_services \
+        "http://${ip_address}:8888/v1/translation" \
+        "choices" \
+        "file-translation" \
+        "polylingua-xeon-backend-server" \
+        '-F "file=@test_data/sample.txt" -F "language_from=English" -F "language_to=Spanish"' \
+        "multipart/form-data"
+}
+
+function validate_nginx() {
+    echo "======================================"
+    echo "Validating Nginx Proxy"
+    echo "======================================"
+
+    # Test translation via nginx
+    validate_services \
+        "http://${ip_address}:80/v1/translation" \
+        "choices" \
+        "nginx-proxy" \
+        "polylingua-nginx-server" \
+        '{"language_from": "English", "language_to": "Italian", "source_language": "Thank you very much"}'
+}
+
+function validate_ui() {
+    echo "======================================"
+    echo "Validating UI Service"
+    echo "======================================"
+
+    # Check if UI is accessible
+    local HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" http://${ip_address}:5173)
+
+    if [ "$HTTP_STATUS" -eq 200 ]; then
+        echo "[ UI ] ✓ UI service is accessible"
+    else
+        echo "[ UI ] ✗ UI service returned HTTP status $HTTP_STATUS"
+        docker logs polylingua-ui-server
+        exit 1
+    fi
+}
+
+function stop_docker() {
+    cd $WORKPATH/docker_compose/intel/cpu/xeon/
+    echo "Stopping services..."
+    docker compose -f compose.yaml down
+    echo "Services stopped"
+}
+
+function main() {
+    echo "======================================"
+    echo "PolyLingua E2E Test Suite"
+    echo "======================================"
+    echo "Platform: Intel Xeon (CPU)"
+    echo "LLM Backend: vLLM"
+    echo "IP Address: ${ip_address}"
+    echo "======================================"
+
+    echo "::group::stop_docker"
+    stop_docker
+    echo "::endgroup::"
+
+    echo "::group::build_docker_images"
+    if [[ "$IMAGE_REPO" == "opea" ]]; then
+        build_docker_images
+    else
+        echo "Skipping image build (using IMAGE_REPO=${IMAGE_REPO})"
+    fi
+    echo "::endgroup::"
+
+    echo "::group::start_services"
+    start_services
+    echo "::endgroup::"
+
+    echo "::group::validate_microservices"
+    validate_microservices
+    echo "::endgroup::"
+
+    echo "::group::validate_megaservice"
+    validate_megaservice
+    echo "::endgroup::"
+
+    echo "::group::validate_file_translation"
+    validate_file_translation
+    echo "::endgroup::"
+
+    echo "::group::validate_nginx"
+    validate_nginx
+    echo "::endgroup::"
+
+    echo "::group::validate_ui"
+    validate_ui
+    echo "::endgroup::"
+
+    echo "::group::stop_docker"
+    stop_docker
+    echo "::endgroup::"
+
+    docker system prune -f
+
+    echo "======================================"
+    echo "✓ All tests passed successfully!"
+    echo "======================================"
+}
+
+main

From 713b55d16a61e22e7c6c5b59ec053f227322c182 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 17 Dec 2025 09:31:05 +0000
Subject: [PATCH 2/9] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 PolyLingua/README.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/PolyLingua/README.md b/PolyLingua/README.md
index 5b5cbba5b4..4b8b5b6198 100644
--- a/PolyLingua/README.md
+++ b/PolyLingua/README.md
@@ -111,7 +111,8 @@ The service works with any HuggingFace text generation model. Recommended models
 
 - **swiss-ai/Apertus-8B-Instruct-2509** - Multilingual translation (default)
 - **haoranxu/ALMA-7B** - Specialized translation model
-- **Qwen/Qwen2.5-7B-Instruct88 - Other Common model (Gaudi default)
+- \*\*Qwen/Qwen2.5-7B-Instruct88 - Other Common model (Gaudi default)
+
 ## 🛠️ Development
 
 ### Project Structure

From 3c6ff746ec2c76c0c8d9b80b5ebc392c772e9e7c Mon Sep 17 00:00:00 2001
From: "Yao, Qing" <qing.yao@intel.com>
Date: Thu, 18 Dec 2025 08:45:24 +0800
Subject: [PATCH 3/9] Fix some typos.

Signed-off-by: Yao, Qing <qing.yao@intel.com>
---
 PolyLingua/README.md                                 |  2 +-
 .../docker_compose/intel/hpu/gaudi/compose.yaml      |  8 ++++----
 PolyLingua/tests/test_compose_on_gaudi.sh            | 12 ++++++------
 3 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/PolyLingua/README.md b/PolyLingua/README.md
index 4b8b5b6198..1c06e9abdf 100644
--- a/PolyLingua/README.md
+++ b/PolyLingua/README.md
@@ -111,7 +111,7 @@ The service works with any HuggingFace text generation model. Recommended models
 
 - **swiss-ai/Apertus-8B-Instruct-2509** - Multilingual translation (default)
 - **haoranxu/ALMA-7B** - Specialized translation model
-- \*\*Qwen/Qwen2.5-7B-Instruct88 - Other Common model (Gaudi default)
+- \*\*Qwen/Qwen2.5-7B-Instruct - Other Common model (Gaudi default)
 
 ## 🛠️ Development
 
diff --git a/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
index 5878d36fb8..3c2e63f0cf 100644
--- a/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
+++ b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
@@ -50,9 +50,9 @@ services:
       HF_HUB_ENABLE_HF_TRANSFER: 0
     restart: unless-stopped
 
-  polylingua-xeon-backend-server:
+  polylingua-gaudi-backend-server:
     image: ${REGISTRY:-opea}/polylingua:${TAG:-latest}
-    container_name: polylingua-xeon-backend-server
+    container_name: polylingua-gaudi-backend-server
     depends_on:
       - vllm-service
       - llm
@@ -73,7 +73,7 @@ services:
     image: ${REGISTRY:-opea}/polylingua-ui:${TAG:-latest}
     container_name: polylingua-ui-server
     depends_on:
-      - polylingua-xeon-backend-server
+      - polylingua-gaudi-backend-server
     ports:
       - "5173:5173"
     environment:
@@ -88,7 +88,7 @@ services:
     image: nginx:alpine
     container_name: polylingua-nginx-server
     depends_on:
-      - polylingua-xeon-backend-server
+      - polylingua-gaudi-backend-server
       - polylingua-ui-server
     ports:
       - "${NGINX_PORT:-80}:80"
diff --git a/PolyLingua/tests/test_compose_on_gaudi.sh b/PolyLingua/tests/test_compose_on_gaudi.sh
index 976d1575f1..4add164694 100755
--- a/PolyLingua/tests/test_compose_on_gaudi.sh
+++ b/PolyLingua/tests/test_compose_on_gaudi.sh
@@ -5,11 +5,11 @@
 set -xe
 IMAGE_REPO=${IMAGE_REPO:-"opea"}
 IMAGE_TAG=${IMAGE_TAG:-"latest"}
-echo "REGISTRY=IMAGE_REPO=${IMAGE_REPO}"
-echo "TAG=IMAGE_TAG=${IMAGE_TAG}"
+echo "IMAGE_REPO=${IMAGE_REPO}"
+echo "IMAGE_TAG=${IMAGE_TAG}"
 export REGISTRY=${IMAGE_REPO}
 export TAG=${IMAGE_TAG}
-export MODEL_CACHE=${model_cache:-"./data"}
+export MODEL_CACHE=${MODEL_CACHE:-"./data"}
 
 # Get the directory where this script is located
 SCRIPT_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
@@ -259,7 +259,7 @@ function validate_file_translation() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "file-translation" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-gaudi-backend-server" \
         '-F "file=@test_data/sample.txt" -F "language_from=English" -F "language_to=Spanish"' \
         "multipart/form-data"
 }
@@ -296,7 +296,7 @@ function validate_ui() {
 }
 
 function stop_docker() {
-    cd $WORKPATH/docker_compose/intel/cpu/xeon/
+    cd $WORKPATH/docker_compose/intel/hpu/gaudi/
     echo "Stopping services..."
     docker compose -f compose.yaml down
     echo "Services stopped"
@@ -306,7 +306,7 @@ function main() {
     echo "======================================"
     echo "PolyLingua E2E Test Suite"
     echo "======================================"
-    echo "Platform: Intel Xeon (CPU)"
+    echo "Platform: Intel Gaudi (HPU)"
     echo "LLM Backend: vLLM"
     echo "IP Address: ${ip_address}"
     echo "======================================"

From cefa5df55c0055337d6c44bb38212ef472b67aa0 Mon Sep 17 00:00:00 2001
From: "Yao, Qing" <qing.yao@intel.com>
Date: Thu, 18 Dec 2025 09:40:50 +0800
Subject: [PATCH 4/9] Fix some typos #2.

Signed-off-by: Yao, Qing <qing.yao@intel.com>
---
 PolyLingua/tests/test_compose_on_gaudi.sh | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/PolyLingua/tests/test_compose_on_gaudi.sh b/PolyLingua/tests/test_compose_on_gaudi.sh
index 4add164694..c72cd2f48e 100755
--- a/PolyLingua/tests/test_compose_on_gaudi.sh
+++ b/PolyLingua/tests/test_compose_on_gaudi.sh
@@ -96,7 +96,7 @@ EOF
     echo "Waiting for vLLM service to initialize (this may take several minutes)..."
     n=0
     until [[ "$n" -ge 100 ]]; do
-        docker logs vllm-service > ${LOG_PATH}/vllm_service_start.log 2>&1
+        docker logs vllm-gaudi-server > ${LOG_PATH}/vllm_service_start.log 2>&1
         if grep -E "Uvicorn running|Application startup complete" ${LOG_PATH}/vllm_service_start.log; then
             echo "vLLM service is ready!"
             break
@@ -112,7 +112,7 @@ EOF
 
     if [[ "$n" -ge 100 ]]; then
         echo "Timeout waiting for vLLM service"
-        docker logs vllm-service
+        docker logs vllm-gaudi-server
         exit 1
     fi
 
@@ -198,7 +198,7 @@ function validate_microservices() {
         "http://${ip_address}:8028/v1/chat/completions" \
         "content" \
         "vllm" \
-        "vllm-service" \
+        "vllm-gaudi-server" \
         '{"model": "Qwen/Qwen2.5-7B-Instruct", "messages": [{"role": "user", "content": "Translate Hello to Spanish"}], "max_tokens": 32}'
 
     # Test LLM microservice
@@ -221,7 +221,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-basic" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-gaudi-backend-server" \
         '{"language_from": "English", "language_to": "Spanish", "source_language": "Hello, how are you today?"}'
 
     # Test 2: Language auto-detection
@@ -230,7 +230,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-auto" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-gaudi-backend-server" \
         '{"language_from": "auto", "language_to": "French", "source_language": "Hello world"}'
 
     # Test 3: Different language pair (English to German)
@@ -239,7 +239,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-german" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-gaudi-backend-server" \
         '{"language_from": "English", "language_to": "German", "source_language": "Good morning"}'
 }
 

From d0090f485be0ea47ba64b40420e36bc06803a749 Mon Sep 17 00:00:00 2001
From: "Yao, Qing" <qing.yao@intel.com>
Date: Thu, 18 Dec 2025 10:11:10 +0800
Subject: [PATCH 5/9] Update vllm gaudi version.

Signed-off-by: Yao, Qing <qing.yao@intel.com>
---
 PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
index 3c2e63f0cf..c626a33b11 100644
--- a/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
+++ b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
@@ -3,7 +3,7 @@
 
 services:
   vllm-service:
-    image: opea/vllm-gaudi:1.4
+    image: opea/vllm-gaudi:1.22.0
     container_name: vllm-gaudi-server
     ports:
       - "8028:80"

From 15092123b0946d74a6a0f60dd7a5af553724b644 Mon Sep 17 00:00:00 2001
From: "Yao, Qing" <qing.yao@intel.com>
Date: Thu, 18 Dec 2025 10:36:54 +0800
Subject: [PATCH 6/9] Fix polylingua-backend container name

Signed-off-by: Yao, Qing <qing.yao@intel.com>
---
 PolyLingua/README.md                                   | 2 +-
 PolyLingua/deploy/nginx.conf                           | 2 +-
 PolyLingua/docker_compose/intel/cpu/xeon/compose.yaml  | 8 ++++----
 PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml | 8 ++++----
 PolyLingua/tests/test_compose_on_gaudi.sh              | 8 ++++----
 PolyLingua/tests/test_compose_on_xeon.sh               | 8 ++++----
 6 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/PolyLingua/README.md b/PolyLingua/README.md
index 1c06e9abdf..22f48a4ee6 100644
--- a/PolyLingua/README.md
+++ b/PolyLingua/README.md
@@ -207,7 +207,7 @@ Translate text between languages.
 docker compose logs -f
 
 # Specific service
-docker compose logs -f polylingua-xeon-backend-server
+docker compose logs -f polylingua-backend-server
 docker compose logs -f polylingua-ui-server
 ```
 
diff --git a/PolyLingua/deploy/nginx.conf b/PolyLingua/deploy/nginx.conf
index 7476021049..cfc152eab4 100644
--- a/PolyLingua/deploy/nginx.conf
+++ b/PolyLingua/deploy/nginx.conf
@@ -26,7 +26,7 @@ http {
 
     # Backend server
     upstream backend {
-        server polylingua-xeon-backend-server:8888;
+        server polylingua-backend-server:8888;
     }
 
     server {
diff --git a/PolyLingua/docker_compose/intel/cpu/xeon/compose.yaml b/PolyLingua/docker_compose/intel/cpu/xeon/compose.yaml
index 5741acadc6..3aa14418b2 100644
--- a/PolyLingua/docker_compose/intel/cpu/xeon/compose.yaml
+++ b/PolyLingua/docker_compose/intel/cpu/xeon/compose.yaml
@@ -45,9 +45,9 @@ services:
       HF_HUB_ENABLE_HF_TRANSFER: 0
     restart: unless-stopped
 
-  polylingua-xeon-backend-server:
+  polylingua-backend-server:
     image: ${REGISTRY:-opea}/polylingua:${TAG:-latest}
-    container_name: polylingua-xeon-backend-server
+    container_name: polylingua-backend-server
     depends_on:
       - vllm-service
       - llm
@@ -68,7 +68,7 @@ services:
     image: ${REGISTRY:-opea}/polylingua-ui:${TAG:-latest}
     container_name: polylingua-ui-server
     depends_on:
-      - polylingua-xeon-backend-server
+      - polylingua-backend-server
     ports:
       - "5173:5173"
     environment:
@@ -83,7 +83,7 @@ services:
     image: nginx:alpine
     container_name: polylingua-nginx-server
     depends_on:
-      - polylingua-xeon-backend-server
+      - polylingua-backend-server
       - polylingua-ui-server
     ports:
       - "${NGINX_PORT:-80}:80"
diff --git a/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
index c626a33b11..6d3ff9bd42 100644
--- a/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
+++ b/PolyLingua/docker_compose/intel/hpu/gaudi/compose.yaml
@@ -50,9 +50,9 @@ services:
       HF_HUB_ENABLE_HF_TRANSFER: 0
     restart: unless-stopped
 
-  polylingua-gaudi-backend-server:
+  polylingua-backend-server:
     image: ${REGISTRY:-opea}/polylingua:${TAG:-latest}
-    container_name: polylingua-gaudi-backend-server
+    container_name: polylingua-backend-server
     depends_on:
       - vllm-service
       - llm
@@ -73,7 +73,7 @@ services:
     image: ${REGISTRY:-opea}/polylingua-ui:${TAG:-latest}
     container_name: polylingua-ui-server
     depends_on:
-      - polylingua-gaudi-backend-server
+      - polylingua-backend-server
     ports:
       - "5173:5173"
     environment:
@@ -88,7 +88,7 @@ services:
     image: nginx:alpine
     container_name: polylingua-nginx-server
     depends_on:
-      - polylingua-gaudi-backend-server
+      - polylingua-backend-server
       - polylingua-ui-server
     ports:
       - "${NGINX_PORT:-80}:80"
diff --git a/PolyLingua/tests/test_compose_on_gaudi.sh b/PolyLingua/tests/test_compose_on_gaudi.sh
index c72cd2f48e..4ac690018a 100755
--- a/PolyLingua/tests/test_compose_on_gaudi.sh
+++ b/PolyLingua/tests/test_compose_on_gaudi.sh
@@ -221,7 +221,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-basic" \
-        "polylingua-gaudi-backend-server" \
+        "polylingua-backend-server" \
         '{"language_from": "English", "language_to": "Spanish", "source_language": "Hello, how are you today?"}'
 
     # Test 2: Language auto-detection
@@ -230,7 +230,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-auto" \
-        "polylingua-gaudi-backend-server" \
+        "polylingua-backend-server" \
         '{"language_from": "auto", "language_to": "French", "source_language": "Hello world"}'
 
     # Test 3: Different language pair (English to German)
@@ -239,7 +239,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-german" \
-        "polylingua-gaudi-backend-server" \
+        "polylingua-backend-server" \
         '{"language_from": "English", "language_to": "German", "source_language": "Good morning"}'
 }
 
@@ -259,7 +259,7 @@ function validate_file_translation() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "file-translation" \
-        "polylingua-gaudi-backend-server" \
+        "polylingua-backend-server" \
         '-F "file=@test_data/sample.txt" -F "language_from=English" -F "language_to=Spanish"' \
         "multipart/form-data"
 }
diff --git a/PolyLingua/tests/test_compose_on_xeon.sh b/PolyLingua/tests/test_compose_on_xeon.sh
index 7bcd67496f..282300f9b5 100755
--- a/PolyLingua/tests/test_compose_on_xeon.sh
+++ b/PolyLingua/tests/test_compose_on_xeon.sh
@@ -220,7 +220,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-basic" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-backend-server" \
         '{"language_from": "English", "language_to": "Spanish", "source_language": "Hello, how are you today?"}'
 
     # Test 2: Language auto-detection
@@ -229,7 +229,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-auto" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-backend-server" \
         '{"language_from": "auto", "language_to": "French", "source_language": "Hello world"}'
 
     # Test 3: Different language pair (English to German)
@@ -238,7 +238,7 @@ function validate_megaservice() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "mega-polylingua-german" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-backend-server" \
         '{"language_from": "English", "language_to": "German", "source_language": "Good morning"}'
 }
 
@@ -258,7 +258,7 @@ function validate_file_translation() {
         "http://${ip_address}:8888/v1/translation" \
         "choices" \
         "file-translation" \
-        "polylingua-xeon-backend-server" \
+        "polylingua-backend-server" \
         '-F "file=@test_data/sample.txt" -F "language_from=English" -F "language_to=Spanish"' \
         "multipart/form-data"
 }

From 6397780aca8ff5caaef5f71516e41f2b1cf5b9e1 Mon Sep 17 00:00:00 2001
From: "Yao, Qing" <qing.yao@intel.com>
Date: Thu, 18 Dec 2025 13:12:57 +0800
Subject: [PATCH 7/9] For CI debug

Signed-off-by: Yao, Qing <qing.yao@intel.com>
---
 PolyLingua/tests/test_compose_on_gaudi.sh | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/PolyLingua/tests/test_compose_on_gaudi.sh b/PolyLingua/tests/test_compose_on_gaudi.sh
index 4ac690018a..50a6e2b564 100755
--- a/PolyLingua/tests/test_compose_on_gaudi.sh
+++ b/PolyLingua/tests/test_compose_on_gaudi.sh
@@ -90,7 +90,17 @@ EOF
 
     # Start Docker Containers
     echo "Starting services with docker compose..."
-    docker compose -f compose.yaml up -d > ${LOG_PATH}/start_services_with_compose.log 2>&1
+    if ! docker compose -f compose.yaml up -d > ${LOG_PATH}/start_services_with_compose.log 2>&1; then
+        echo "::error::Docker Compose failed to start. Printing logs..."
+        cat "${LOG_PATH}/start_services_with_compose.log"
+        echo "::group::Docker Compose PS"
+        docker compose -f compose.yaml ps
+        echo "::endgroup::"
+        echo "::group::Docker Logs"
+        docker compose -f compose.yaml logs
+        echo "::endgroup::"
+        exit 1
+    fi
 
     # Wait for vLLM service to be ready
     echo "Waiting for vLLM service to initialize (this may take several minutes)..."

From d4af50c67e7c642fd805646db2b15bc5c4145e08 Mon Sep 17 00:00:00 2001
From: "Yao, Qing" <qing.yao@intel.com>
Date: Thu, 18 Dec 2025 14:10:43 +0800
Subject: [PATCH 8/9] For CI debug #2

Signed-off-by: Yao, Qing <qing.yao@intel.com>
---
 PolyLingua/tests/test_compose_on_gaudi.sh | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/PolyLingua/tests/test_compose_on_gaudi.sh b/PolyLingua/tests/test_compose_on_gaudi.sh
index 50a6e2b564..4a424481f2 100755
--- a/PolyLingua/tests/test_compose_on_gaudi.sh
+++ b/PolyLingua/tests/test_compose_on_gaudi.sh
@@ -35,9 +35,20 @@ function build_docker_images() {
     # Build all images using build.yaml
     echo "Building PolyLingua images with --no-cache, check docker_image_build.log for details..."
     service_list="polylingua polylingua-ui llm-textgen"
-    docker compose -f build.yaml build ${service_list} --no-cache > ${LOG_PATH}/docker_image_build.log 2>&1
+
+    if ! docker compose -f build.yaml build ${service_list} --no-cache > ${LOG_PATH}/docker_image_build.log 2>&1; then
+        echo "::error::Docker Compose build failed. Printing build logs..."
+        cat "${LOG_PATH}/docker_image_build.log"
+        exit 1
+    fi
 
     echo "Image build completed"
+    echo "Verifying built images..."
+    if ! docker images | grep -q "polylingua" || ! docker images | grep -q "polylingua-ui" || ! docker images | grep -q "llm-textgen"; then
+        echo "::error::One or more required images are missing after build!"
+        docker images
+        exit 1
+    fi
     docker images | grep -E "polylingua|llm-textgen"
     sleep 1s
 }

From 51ba6d632d01f7c4c0736d29384aab20d52e78ae Mon Sep 17 00:00:00 2001
From: "Yao, Qing" <qing.yao@intel.com>
Date: Thu, 18 Dec 2025 14:20:42 +0800
Subject: [PATCH 9/9] For CI debug #3

Signed-off-by: Yao, Qing <qing.yao@intel.com>
---
 PolyLingua/tests/test_compose_on_gaudi.sh | 28 +++++++++++++++--------
 1 file changed, 18 insertions(+), 10 deletions(-)

diff --git a/PolyLingua/tests/test_compose_on_gaudi.sh b/PolyLingua/tests/test_compose_on_gaudi.sh
index 4a424481f2..4b4158b420 100755
--- a/PolyLingua/tests/test_compose_on_gaudi.sh
+++ b/PolyLingua/tests/test_compose_on_gaudi.sh
@@ -114,25 +114,33 @@ EOF
     fi
 
     # Wait for vLLM service to be ready
-    echo "Waiting for vLLM service to initialize (this may take several minutes)..."
+    echo "Waiting for vLLM service to become healthy (this may take up to 30 minutes)..."
+    local vllm_health_url="http://${ip_address}:8028/health"
+
     n=0
-    until [[ "$n" -ge 100 ]]; do
-        docker logs vllm-gaudi-server > ${LOG_PATH}/vllm_service_start.log 2>&1
-        if grep -E "Uvicorn running|Application startup complete" ${LOG_PATH}/vllm_service_start.log; then
-            echo "vLLM service is ready!"
+    until [[ "$n" -ge 180 ]]; do
+        http_status=$(curl -s -o /dev/null -w "%{http_code}" "$vllm_health_url")
+
+        if [[ "$http_status" -eq 200 ]]; then
+            echo "vLLM service is healthy (returned status 200)!"
             break
         fi
-        if grep -q "error" ${LOG_PATH}/vllm_service_start.log; then
-            echo "Error detected in vLLM service startup"
-            cat ${LOG_PATH}/vllm_service_start.log
+
+        if ! docker ps --filter "name=vllm-gaudi-server" --filter "status=running" -q | grep -q .; then
+            echo "::error::vLLM container has stopped unexpectedly!"
+            docker logs vllm-gaudi-server
             exit 1
         fi
+
+        echo "Waiting for vLLM health endpoint... status: $http_status ($n/180)"
         sleep 10s
         n=$((n+1))
     done
 
-    if [[ "$n" -ge 100 ]]; then
-        echo "Timeout waiting for vLLM service"
+    if [[ "$n" -ge 180 ]]; then
+        echo "::error::Timeout waiting for vLLM service after 30 minutes."
+        echo "Final health check status: $http_status"
+        echo "Dumping container logs:"
         docker logs vllm-gaudi-server
         exit 1
     fi