Adding files to deploy Translation application on ROCm vLLM (#1648)

Signed-off-by: Chingis Yundunov <YundunovCN@sibedge.com> Signed-off-by: Artem Astafev <a.astafev@datamonsters.com>
2025-03-31 12:49:33 +07:00
parent b7f24762a3
commit 340fa075bd
8 changed files with 704 additions and 58 deletions
--- a/Translation/assets/img/translation-ui-response-example.png
+++ b/Translation/assets/img/translation-ui-response-example.png
--- a/Translation/assets/img/translation-ui-starting-page.png
+++ b/Translation/assets/img/translation-ui-starting-page.png
--- a/Translation/docker_compose/amd/gpu/rocm/README.md
+++ b/Translation/docker_compose/amd/gpu/rocm/README.md
@@ -1,42 +1,117 @@
 # Build and deploy Translation Application on AMD GPU (ROCm)

-## Build images
+## Build Docker Images

-### Build the LLM Docker Image
+### 1. Build Docker Image

-```bash
-### Cloning repo
-git clone https://github.com/opea-project/GenAIComps.git
-cd GenAIComps
+- #### Create application install directory and go to it:

-### Build Docker image
-docker build -t opea/llm-textgen:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/llms/src/text-generation/Dockerfile .
-```
+  ```bash
+  mkdir ~/translation-install && cd translation-install
+  ```

-### Build the MegaService Docker Image
+- #### Clone the repository GenAIExamples (the default repository branch "main" is used here):

-```bash
-### Cloning repo
-git clone https://github.com/opea-project/GenAIExamples
-cd GenAIExamples/Translation/
+  ```bash
+  git clone https://github.com/opea-project/GenAIExamples.git
+  ```

-### Build Docker image
-docker build -t opea/translation:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile .
-```
+  If you need to use a specific branch/tag of the GenAIExamples repository, then (v1.3 replace with its own value):

-### Build the UI Docker Image
+  ```bash
+  git clone https://github.com/opea-project/GenAIExamples.git && cd GenAIExamples && git checkout v1.3
+  ```

-```bash
-cd GenAIExamples/Translation/ui
-### Build UI Docker image
-docker build -t opea/translation-ui:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f docker/Dockerfile .
-```
+  We remind you that when using a specific version of the code, you need to use the README from this version:

-## Deploy Translation Application
+- #### Go to build directory:

-### Features of Docker compose for AMD GPUs
+  ```bash
+  cd ~/translation-install/GenAIExamples/Translation/docker_image_build
+  ```

-1. Added forwarding of GPU devices to the container TGI service with instructions:
+- Cleaning up the GenAIComps repository if it was previously cloned in this directory.
+  This is necessary if the build was performed earlier and the GenAIComps folder exists and is not empty:
+
+  ```bash
+  echo Y | rm -R GenAIComps
+  ```
+
+- #### Clone the repository GenAIComps (the default repository branch "main" is used here):
+
+  ```bash
+  git clone https://github.com/opea-project/GenAIComps.git
+  ```
+
+  If you use a specific tag of the GenAIExamples repository,
+  then you should also use the corresponding tag for GenAIComps. (v1.3 replace with its own value):
+
+  ```bash
+  git clone https://github.com/opea-project/GenAIComps.git && cd GenAIComps && git checkout v1.3
+  ```
+
+  We remind you that when using a specific version of the code, you need to use the README from this version.
+
+- #### Setting the list of images for the build (from the build file.yaml)
+
+  If you want to deploy a vLLM-based or TGI-based application, then the set of services is installed as follows:
+
+  #### vLLM-based application
+
+  ```bash
+  service_list="vllm-rocm translation translation-ui llm-textgen nginx"
+  ```
+
+  #### TGI-based application
+
+  ```bash
+  service_list="translation translation-ui llm-textgen nginx"
+  ```
+
+- #### Optional. Pull TGI Docker Image (Do this if you want to use TGI)
+
+  ```bash
+  docker pull ghcr.io/huggingface/text-generation-inference:2.3.1-rocm
+  ```
+
+- #### Build Docker Images
+
+  ```bash
+  docker compose -f build.yaml build ${service_list} --no-cache
+  ```
+
+  After the build, we check the list of images with the command:
+
+  ```bash
+  docker image ls
+  ```
+
+  The list of images should include:
+
+  ##### vLLM-based application:
+
+  - opea/vllm-rocm:latest
+    - opea/llm-textgen:latest
+    - opea/nginx:latest
+    - opea/translation:latest
+    - opea/translation-ui:latest
+
+  ##### TGI-based application:
+
+  - ghcr.io/huggingface/text-generation-inference:2.3.1-rocm
+    - opea/llm-textgen:latest
+    - opea/nginx:latest
+    - opea/translation:latest
+    - opea/translation-ui:latest
+
+---
+
+### Docker Compose Configuration for AMD GPUs
+
+To enable GPU support for AMD GPUs, the following configuration is added to the Docker Compose file:
+
+- compose_vllm.yaml - for vLLM-based application
+- compose.yaml - for TGI-based

 ```yaml
 shm_size: 1g
@@ -51,16 +126,14 @@ security_opt:
  - seccomp:unconfined
 ```

-In this case, all GPUs are thrown. To reset a specific GPU, you need to use specific device names cardN and renderN.
-
-For example:
+This configuration forwards all available GPUs to the container. To use a specific GPU, specify its `cardN` and `renderN` device IDs. For example:

 ```yaml
 shm_size: 1g
 devices:
  - /dev/kfd:/dev/kfd
  - /dev/dri/card0:/dev/dri/card0
-  - /dev/dri/render128:/dev/dri/render128
+  - /dev/dri/renderD128:/dev/dri/renderD128
 cap_add:
  - SYS_PTRACE
 group_add:
@@ -69,60 +142,305 @@ security_opt:
  - seccomp:unconfined
 ```

-To find out which GPU device IDs cardN and renderN correspond to the same GPU, use the GPU driver utility
+**How to Identify GPU Device IDs:**
+Use AMD GPU driver utilities to determine the correct `cardN` and `renderN` IDs for your GPU.

-### Go to the directory with the Docker compose file
+### Set deploy environment variables
+
+#### Setting variables in the operating system environment:
+
+##### Set variable HUGGINGFACEHUB_API_TOKEN:

 ```bash
-cd GenAIExamples/Translation/docker_compose/amd/gpu/rocm
+### Replace the string 'your_huggingfacehub_token' with your HuggingFacehub repository access token.
+export HUGGINGFACEHUB_API_TOKEN='your_huggingfacehub_token'
 ```

-### Set environments
+#### Set variables value in set_env\*\*\*\*.sh file:

-In the file "GenAIExamples/Translation/docker_compose/amd/gpu/rocm/set_env.sh " it is necessary to set the required values. Parameter assignments are specified in the comments for each variable setting command
+Go to Docker Compose directory:
+
+```bash
+cd ~/translation-install/GenAIExamples/Translation/docker_compose/amd/gpu/rocm
+```
+
+The example uses the Nano text editor. You can use any convenient text editor:
+
+#### If you use vLLM
+
+```bash
+nano set_env_vllm.sh
+```
+
+#### If you use TGI
+
+```bash
+nano set_env.sh
+```
+
+If you are in a proxy environment, also set the proxy-related environment variables:
+
+```bash
+export http_proxy="Your_HTTP_Proxy"
+export https_proxy="Your_HTTPs_Proxy"
+```
+
+Set the values of the variables:
+
+- **HOST_IP, HOST_IP_EXTERNAL** - These variables are used to configure the name/address of the service in the operating system environment for the application services to interact with each other and with the outside world.
+
+  If your server uses only an internal address and is not accessible from the Internet, then the values for these two variables will be the same and the value will be equal to the server's internal name/address.
+
+  If your server uses only an external, Internet-accessible address, then the values for these two variables will be the same and the value will be equal to the server's external name/address.
+
+  If your server is located on an internal network, has an internal address, but is accessible from the Internet via a proxy/firewall/load balancer, then the HOST_IP variable will have a value equal to the internal name/address of the server, and the EXTERNAL_HOST_IP variable will have a value equal to the external name/address of the proxy/firewall/load balancer behind which the server is located.
+
+  We set these values in the file set_env\*\*\*\*.sh
+
+- **Variables with names like "**\*\*\*\*\*\*\_PORT"\*\* - These variables set the IP port numbers for establishing network connections to the application services.
+  The values shown in the file set_env.sh or set_env_vllm they are the values used for the development and testing of the application, as well as configured for the environment in which the development is performed. These values must be configured in accordance with the rules of network access to your environment's server, and must not overlap with the IP ports of other applications that are already in use.
+
+#### Set variables with script set_env\*\*\*\*.sh
+
+#### If you use vLLM
+
+```bash
+. set_env_vllm.sh
+```
+
+#### If you use TGI

 ```bash
-chmod +x set_env.sh
 . set_env.sh
 ```

-### Run services
+### Start the services:

-```
-docker compose up -d
-```
-
-# Validate the MicroServices and MegaService
-
-## Validate TGI service
+#### If you use vLLM

 ```bash
-curl http://${TRANSLATION_HOST_IP}:${TRANSLATIONS_TGI_SERVICE_PORT}/generate \
+docker compose -f compose_vllm.yaml up -d
+```
+
+#### If you use TGI
+
+```bash
+docker compose -f compose.yaml up -d
+```
+
+All containers should be running and should not restart:
+
+##### If you use vLLM:
+
+- translationn-vllm-service
+- translation-tgi-service
+- translation-llm
+- translation-backend-server
+- translation-ui-server
+- translation-nginx-server
+
+##### If you use TGI:
+
+- translation-tgi-service
+- translation-llm
+- translation-backend-server
+- translation-ui-server
+- translation-nginx-server
+
+---
+
+## Validate the Services
+
+### 1. Validate the vLLM/TGI Service
+
+#### If you use vLLM:
+
+```bash
+DATA='{"model": "haoranxu/ALMA-13B", "prompt": "What is Deep Learning?", "max_tokens": 100, "temperature": 0}'
+
+curl http://${HOST_IP}:${TRANSLATION_VLLM_SERVICE_PORT}/v1/chat/completions \
  -X POST \
-  -d '{"inputs":"What is Deep Learning?","parameters":{"max_new_tokens":17, "do_sample": true}}' \
+  -d "$DATA" \
  -H 'Content-Type: application/json'
 ```

-## Validate LLM service
+Checking the response from the service. The response should be similar to JSON:
+
+```json
+{
+  "id": "cmpl-059dd7fb311a46c2b807e0b3315e730c",
+  "object": "text_completion",
+  "created": 1743063706,
+  "model": "haoranxu/ALMA-13B",
+  "choices": [
+    {
+      "index": 0,
+      "text": " Deep Learning is a subset of machine learning. It attempts to mimic the way the human brain learns. Deep Learning is a subset of machine learning. It attempts to mimic the way the human brain learns. Deep Learning is a subset of machine learning. It attempts to mimic the way the human brain learns. Deep Learning is a subset of machine learning. It attempts to mimic the way the human brain learns. Deep Learning is a subset of machine learning",
+      "logprobs": null,
+      "finish_reason": "length",
+      "stop_reason": null,
+      "prompt_logprobs": null
+    }
+  ],
+
+  "usage": {
+    "prompt_tokens": 6,
+    "total_tokens": 106,
+    "completion_tokens": 100,
+    "prompt_tokens_details": null
+  }
+}
+```
+
+If the service response has a meaningful response in the value of the "choices.message.content" key,
+then we consider the vLLM service to be successfully launched
+
+#### If you use TGI:

 ```bash
-curl http://${TRANSLATION_HOST_IP}:9000/v1/chat/completions \
+DATA='{"inputs":"What is Deep Learning?",'\
+'"parameters":{"max_new_tokens":256,"do_sample": true}}'
+
+curl http://${HOST_IP}:${TRANSLATION_TGI_SERVICE_PORT}/generate \
  -X POST \
-  -d '{"query":"Translate this from Chinese to English:\nChinese: 我爱机器翻译。\nEnglish:"}' \
+  -d "$DATA" \
  -H 'Content-Type: application/json'
 ```

-## Validate MegaService
+Checking the response from the service. The response should be similar to JSON:

-```bash
-curl http://${TRANSLATION_HOST_IP}:${TRANSLATION_BACKEND_SERVICE_PORT}/v1/translation -H "Content-Type: application/json" -d '{
-     "language_from": "Chinese","language_to": "English","source_language": "我爱机器翻译。"}'
+```json
+{
+  "generated_text": "\n\n What can it Do? What's the Hype? What Should You Do If"
+}
 ```

-## Validate Nginx service
+If the service response has a meaningful response in the value of the "generated_text" key,
+then we consider the TGI service to be successfully launched
+
+### 2. Validate the LLM Service

 ```bash
-curl http://${TRANSLATION_HOST_IP}:${TRANSLATION_NGINX_PORT}/v1/translation \
-    -H "Content-Type: application/json" \
-    -d '{"language_from": "Chinese","language_to": "English","source_language": "我爱机器翻译。"}'
+DATA='{"query":"What is Deep Learning?",'\
+'"max_tokens":32,"top_k":10,"top_p":0.95,"typical_p":0.95,"temperature":0.01,'\
+'"repetition_penalty":1.03,"stream":false}'
+
+curl http://${HOST_IP}:${TRANSLATION_LLM_SERVICE_PORT}/v1/chat/completions \
+  -X POST \
+  -d "$DATA" \
+  -H 'Content-Type: application/json'
+```
+
+Checking the response from the service. The response should be similar to JSON:
+
+```json
+{
+  "id": "",
+  "choices": [
+    {
+      "finish_reason": "length",
+      "index": 0,
+      "logprobs": null,
+      "text": " Deep Learning is a subset of machine learning. It attempts to mimic the way the human brain learns. Deep Learning is a subset of machine learning."
+    }
+  ],
+  "created": 1742978568,
+  "model": "haoranxu/ALMA-13B",
+  "object": "text_completion",
+  "system_fingerprint": "2.3.1-sha-a094729-rocm",
+  "usage": {
+    "completion_tokens": 32,
+    "prompt_tokens": 6,
+    "total_tokens": 38,
+    "completion_tokens_details": null,
+    "prompt_tokens_details": null
+  }
+}
+```
+
+### 3. Validate Nginx Service
+
+```bash
+DATA='{"language_from": "Chinese","language_to": "English","source_language": "我爱机器翻译。"}'
+
+curl http://${HOST_IP}:${TRANSLATION_LLM_SERVICE_PORT}/v1/translation  \
+  -d "$DATA" \
+  -H 'Content-Type: application/json'
+```
+
+Checking the response from the service. The response should be similar to JSON:
+
+```textmate
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" I"}],"created":1743062099,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" love"}],"created":1743062099,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" machine"}],"created":1743062099,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" translation"}],"created":1743062099,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":"."}],"created":1743062099,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"eos_token","index":0,"logprobs":null,"text":"</s>"}],"created":1743062099,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":{"completion_tokens":6,"prompt_tokens":3071,"total_tokens":3077,"completion_tokens_details":null,"prompt_tokens_details":null}}
+
+data: [DONE]
+```
+
+### 4. Validate MegaService
+
+```bash
+DATA='{"language_from": "Chinese","language_to": "English","source_language": "我爱机器翻译。"}'
+
+curl http://${HOST_IP}:${TRANSLATION_BACKEND_SERVICE_PORT}/v1/translation \
+  -H "Content-Type: application/json" \
+  -d "$DATA"
+```
+
+Checking the response from the service. The response should be similar to JSON:
+
+```textmate
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" I"}],"created":1742978968,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" love"}],"created":1742978968,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" machine"}],"created":1742978968,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":" translation"}],"created":1742978968,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"","index":0,"logprobs":null,"text":"."}],"created":1742978968,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":null}
+
+data: {"id":"","choices":[{"finish_reason":"eos_token","index":0,"logprobs":null,"text":"</s>"}],"created":1742978968,"model":"haoranxu/ALMA-13B","object":"text_completion","system_fingerprint":"2.3.1-sha-a094729-rocm","usage":{"completion_tokens":6,"prompt_tokens":3071,"total_tokens":3077,"completion_tokens_details":null,"prompt_tokens_details":null}}
+
+data: [DONE]
+
+```
+
+If the response text is similar to the one above, then we consider the service verification successful.
+
+### 5. Validate Frontend
+
+To access the UI, use the URL - http://${EXTERNAL_HOST_IP}:${TRANSLATION_FRONTEND_SERVICE_PORT} A page should open when you click through to this address:
+![UI start page](../../../../assets/img/translation-ui-starting-page.png)
+
+If a page of this type has opened, then we believe that the service is running and responding, and we can proceed to functional UI testing.
+
+Let's enter the task for the service in the "Input" field. For example, "我爱机器翻译" with selected "German" as language source and press Enter. After that, a page with the result of the task should open:
+
+![UI start page](../../../../assets/img/translation-ui-response-example.png)
+If the result shown on the page is correct, then we consider the verification of the UI service to be successful.
+
+### 6. Stop application
+
+#### If you use vLLM
+
+```bash
+cd ~/translation-install/GenAIExamples/Translation/docker_compose/amd/gpu/rocm
+docker compose -f compose_vllm.yaml down
+```
+
+#### If you use TGI
+
+```bash
+cd ~/translation-install/GenAIExamples/Translation/docker_compose/amd/gpu/rocm
+docker compose -f compose.yaml down
 ```
--- a/Translation/docker_compose/amd/gpu/rocm/compose.yaml
+++ b/Translation/docker_compose/amd/gpu/rocm/compose.yaml
@@ -90,7 +90,7 @@ services:
      - translation-backend-server
      - translation-ui-server
    ports:
-      - "${TRANSLATION_NGINX_PORT:-80}:80"
+      - "${TRANSLATION_NGINX_PORT:-80}:8080"
    environment:
      - no_proxy=${no_proxy}
      - https_proxy=${https_proxy}
--- a/Translation/docker_compose/amd/gpu/rocm/compose_vllm.yaml
+++ b/Translation/docker_compose/amd/gpu/rocm/compose_vllm.yaml
@@ -0,0 +1,107 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+services:
+  translation-vllm-service:
+    image: ${REGISTRY:-opea}/vllm-rocm:${TAG:-latest}
+    container_name: translation-vllm-service
+    ports:
+      - "${TRANSLATION_VLLM_SERVICE_PORT:-8081}:8011"
+    environment:
+      no_proxy: ${no_proxy}
+      http_proxy: ${http_proxy}
+      https_proxy: ${https_proxy}
+      HUGGINGFACEHUB_API_TOKEN: ${TRANSLATION_HUGGINGFACEHUB_API_TOKEN}
+      HF_TOKEN: ${TRANSLATION_HUGGINGFACEHUB_API_TOKEN}
+      HF_HUB_DISABLE_PROGRESS_BARS: 1
+      HF_HUB_ENABLE_HF_TRANSFER: 0
+      WILM_USE_TRITON_FLASH_ATTENTION: 0
+      PYTORCH_JIT: 0
+    volumes:
+      - "./data:/data"
+    shm_size: 20G
+    devices:
+      - /dev/kfd:/dev/kfd
+      - /dev/dri/:/dev/dri/
+    cap_add:
+      - SYS_PTRACE
+    group_add:
+      - video
+    security_opt:
+      - seccomp:unconfined
+      - apparmor=unconfined
+    command: "--model ${TRANSLATION_LLM_MODEL_ID} --swap-space 16 --disable-log-requests --dtype float16 --tensor-parallel-size 1 --host 0.0.0.0 --port 8011 --num-scheduler-steps 1 --distributed-executor-backend \"mp\""
+    ipc: host
+  translation-llm:
+    image: ${REGISTRY:-opea}/llm-textgen:${TAG:-latest}
+    container_name: translation-llm-textgen-server
+    depends_on:
+      - translation-vllm-service
+    ports:
+      - "${TRANSLATION_LLM_PORT:-9000}:9000"
+    ipc: host
+    environment:
+      no_proxy: ${no_proxy}
+      http_proxy: ${http_proxy}
+      https_proxy: ${https_proxy}
+      LLM_ENDPOINT: ${TRANSLATION_LLM_ENDPOINT}
+      LLM_MODEL_ID: ${TRANSLATION_LLM_MODEL_ID}
+      HUGGINGFACEHUB_API_TOKEN: ${TRANSLATION_HUGGINGFACEHUB_API_TOKEN}
+      HF_TOKEN: ${TRANSLATION_HUGGINGFACEHUB_API_TOKEN}
+      LLM_COMPONENT_NAME: "OpeaTextGenService"
+      HF_HUB_DISABLE_PROGRESS_BARS: 1
+      HF_HUB_ENABLE_HF_TRANSFER: 0
+    restart: unless-stopped
+  translation-backend-server:
+    image: ${REGISTRY:-opea}/translation:${TAG:-latest}
+    container_name: translation-backend-server
+    depends_on:
+      - translation-vllm-service
+      - translation-llm
+    ports:
+      - "${TRANSLATION_BACKEND_SERVICE_PORT:-8888}:8888"
+    environment:
+      no_proxy: ${no_proxy}
+      https_proxy: ${https_proxy}
+      http_proxy: ${http_proxy}
+      MEGA_SERVICE_HOST_IP: ${TRANSLATION_MEGA_SERVICE_HOST_IP}
+      LLM_SERVICE_HOST_IP: ${TRANSLATION_LLM_SERVICE_HOST_IP}
+      LLM_SERVICE_PORT: ${TRANSLATION_LLM_PORT}
+    ipc: host
+    restart: always
+  translation-ui-server:
+    image: ${REGISTRY:-opea}/translation-ui:${TAG:-latest}
+    container_name: translation-ui-server
+    depends_on:
+      - translation-backend-server
+    ports:
+      - "${TRANSLATION_FRONTEND_SERVICE_PORT:-5173}:5173"
+    environment:
+      no_proxy: ${no_proxy}
+      https_proxy: ${https_proxy}
+      http_proxy: ${http_proxy}
+      BASE_URL: ${TRANSLATION_BACKEND_SERVICE_ENDPOINT}
+    ipc: host
+    restart: always
+  translation-nginx-server:
+    image: ${REGISTRY:-opea}/nginx:${TAG:-latest}
+    container_name: translation-nginx-server
+    depends_on:
+      - translation-backend-server
+      - translation-ui-server
+    ports:
+      - "${TRANSLATION_NGINX_PORT:-80}:8080"
+    environment:
+      no_proxy: ${no_proxy}
+      https_proxy: ${https_proxy}
+      http_proxy: ${http_proxy}
+      FRONTEND_SERVICE_IP: ${TRANSLATION_FRONTEND_SERVICE_IP}
+      FRONTEND_SERVICE_PORT: ${TRANSLATION_FRONTEND_SERVICE_PORT}
+      BACKEND_SERVICE_NAME: ${TRANSLATION_BACKEND_SERVICE_NAME}
+      BACKEND_SERVICE_IP: ${TRANSLATION_BACKEND_SERVICE_IP}
+      BACKEND_SERVICE_PORT: ${TRANSLATION_BACKEND_SERVICE_PORT}
+    ipc: host
+    restart: always
+networks:
+  default:
+    driver: bridge
--- a/Translation/docker_compose/amd/gpu/rocm/set_env_vllm.sh
+++ b/Translation/docker_compose/amd/gpu/rocm/set_env_vllm.sh
@@ -0,0 +1,23 @@
+#!/usr/bin/env bash
+
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+# SPDX-License-Identifier: Apache-2.0
+
+export HOST_IP=''
+export EXTERNAL_HOST_IP=''
+export TRANSLATION_LLM_MODEL_ID="haoranxu/ALMA-13B"
+export TRANSLATION_VLLM_SERVICE_PORT=8088
+export TRANSLATION_LLM_ENDPOINT="http://${HOST_IP}:${TRANSLATION_VLLM_SERVICE_PORT}"
+export TRANSLATION_LLM_PORT=9088
+export TRANSLATION_HUGGINGFACEHUB_API_TOKEN=${HUGGINGFACEHUB_API_TOKEN}
+export TRANSLATION_MEGA_SERVICE_HOST_IP=${HOST_IP}
+export TRANSLATION_LLM_SERVICE_HOST_IP=${HOST_IP}
+export TRANSLATION_FRONTEND_SERVICE_IP=${HOST_IP}
+export TRANSLATION_FRONTEND_SERVICE_PORT=18122
+export TRANSLATION_BACKEND_SERVICE_NAME=translation
+export TRANSLATION_BACKEND_SERVICE_IP=${HOST_IP}
+export TRANSLATION_BACKEND_SERVICE_PORT=18121
+export TRANSLATION_BACKEND_SERVICE_ENDPOINT="http://${EXTERNAL_HOST_IP}:${TRANSLATION_BACKEND_SERVICE_PORT}/v1/translation"
+export TRANSLATION_NGINX_PORT=18123
--- a/Translation/docker_image_build/build.yaml
+++ b/Translation/docker_image_build/build.yaml
@@ -29,3 +29,8 @@ services:
      dockerfile: comps/third_parties/nginx/src/Dockerfile
    extends: translation
    image: ${REGISTRY:-opea}/nginx:${TAG:-latest}
+  vllm-rocm:
+    build:
+      context: GenAIComps
+      dockerfile: comps/third_parties/vllm/src/Dockerfile.amd_gpu
+    image: ${REGISTRY:-opea}/vllm-rocm:${TAG:-latest}
--- a/Translation/tests/test_compose_vllm_on_rocm.sh
+++ b/Translation/tests/test_compose_vllm_on_rocm.sh
@@ -0,0 +1,193 @@
+#!/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}
+
+WORKPATH=$(dirname "$PWD")
+LOG_PATH="$WORKPATH/tests"
+ip_address=$(hostname -I | awk '{print $1}')
+
+function build_docker_images() {
+    opea_branch=${opea_branch:-"main"}
+    # If the opea_branch isn't main, replace the git clone branch in Dockerfile.
+    if [[ "${opea_branch}" != "main" ]]; then
+        cd $WORKPATH
+        OLD_STRING="RUN git clone --depth 1 https://github.com/opea-project/GenAIComps.git"
+        NEW_STRING="RUN git clone --depth 1 --branch ${opea_branch} https://github.com/opea-project/GenAIComps.git"
+        find . -type f -name "Dockerfile*" | while read -r file; do
+            echo "Processing file: $file"
+            sed -i "s|$OLD_STRING|$NEW_STRING|g" "$file"
+        done
+    fi
+
+    cd $WORKPATH/docker_image_build
+    git clone --depth 1 --branch ${opea_branch} https://github.com/opea-project/GenAIComps.git
+
+    echo "Build all the images with --no-cache, check docker_image_build.log for details..."
+    service_list="translation translation-ui llm-textgen nginx vllm-rocm"
+    docker compose -f build.yaml build ${service_list} --no-cache > ${LOG_PATH}/docker_image_build.log
+    docker images && sleep 3s
+}
+
+function start_services() {
+    cd $WORKPATH/docker_compose/amd/gpu/rocm/
+
+    export HOST_IP=${ip_address}
+    export EXTERNAL_HOST_IP=${ip_address}
+    export TRANSLATION_LLM_MODEL_ID="haoranxu/ALMA-13B"
+    export TRANSLATION_VLLM_SERVICE_PORT=8088
+    export TRANSLATION_LLM_ENDPOINT="http://${HOST_IP}:${TRANSLATION_VLLM_SERVICE_PORT}"
+    export TRANSLATION_LLM_PORT=9088
+    export TRANSLATION_HUGGINGFACEHUB_API_TOKEN=${HUGGINGFACEHUB_API_TOKEN}
+    export TRANSLATION_MEGA_SERVICE_HOST_IP=${HOST_IP}
+    export TRANSLATION_LLM_SERVICE_HOST_IP=${HOST_IP}
+    export TRANSLATION_FRONTEND_SERVICE_IP=${HOST_IP}
+    export TRANSLATION_FRONTEND_SERVICE_PORT=5173
+    export TRANSLATION_BACKEND_SERVICE_NAME=translation
+    export TRANSLATION_BACKEND_SERVICE_IP=${HOST_IP}
+    export TRANSLATION_BACKEND_SERVICE_PORT=8089
+    export TRANSLATION_BACKEND_SERVICE_ENDPOINT="http://${EXTERNAL_HOST_IP}:${TRANSLATION_BACKEND_SERVICE_PORT}/v1/translation"
+    export TRANSLATION_NGINX_PORT=8090
+
+    sed -i "s/backend_address/$ip_address/g" $WORKPATH/ui/svelte/.env
+
+    # Start Docker Containers
+    docker compose -f compose_vllm.yaml up -d > ${LOG_PATH}/start_services_with_compose.log
+
+    n=0
+    # wait long for llm model download
+    until [[ "$n" -ge 500 ]]; do
+        docker logs translation-vllm-service >& ${LOG_PATH}/translation-vllm-service_start.log
+        if grep -q "Application startup complete" ${LOG_PATH}/translation-vllm-service_start.log; then
+            echo "vLLM check successful"
+            break
+        fi
+        sleep 10s
+        n=$((n+1))
+    done
+
+}
+
+function validate_services() {
+    local URL="$1"
+    local EXPECTED_RESULT="$2"
+    local SERVICE_NAME="$3"
+    local DOCKER_NAME="$4"
+    local INPUT_DATA="$5"
+
+    local HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" -X POST -d "$INPUT_DATA" -H 'Content-Type: application/json' "$URL")
+    if [ "$HTTP_STATUS" -eq 200 ]; then
+        echo "[ $SERVICE_NAME ] HTTP status is 200. Checking content..."
+
+        local CONTENT=$(curl -s -X POST -d "$INPUT_DATA" -H 'Content-Type: application/json' "$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 the expected result: $CONTENT"
+            docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+            exit 1
+        fi
+    else
+        echo "[ $SERVICE_NAME ] HTTP status is not 200. Received status was $HTTP_STATUS"
+        docker logs ${DOCKER_NAME} >> ${LOG_PATH}/${SERVICE_NAME}.log
+        exit 1
+    fi
+    sleep 1s
+}
+
+function validate_microservices() {
+    # Check if the microservices are running correctly.
+
+    # vLLM for llm service
+    validate_services \
+        "${ip_address}:${TRANSLATION_VLLM_SERVICE_PORT}/v1/completions" \
+        "choices" \
+        "translation-vllm-service" \
+        "translation-vllm-service" \
+        '{"model": "haoranxu/ALMA-13B", "prompt": "What is Deep Learning?", "max_tokens": 100, "temperature": 0}'
+
+    # llm microservice
+    validate_services \
+        "${HOST_IP}:${TRANSLATION_LLM_PORT}/v1/chat/completions" \
+        "data: " \
+        "translation-llm" \
+        "translation-llm-textgen-server" \
+        '{"query":"Translate this from Chinese to English:\nChinese: 我爱机器翻译。\nEnglish:"}'
+}
+
+function validate_megaservice() {
+    # Curl the Mega Service
+    validate_services \
+        "${HOST_IP}:${TRANSLATION_BACKEND_SERVICE_PORT}/v1/translation" \
+        "translation" \
+        "translation-backend-server" \
+        "translation-backend-server" \
+        '{"language_from": "Chinese","language_to": "English","source_language": "我爱机器翻译。"}'
+
+    # test the megeservice via nginx
+    validate_services \
+        "${HOST_IP}:${TRANSLATION_NGINX_PORT}/v1/translation" \
+        "translation" \
+        "translation-nginx-server" \
+        "translation-nginx-server" \
+        '{"language_from": "Chinese","language_to": "English","source_language": "我爱机器翻译。"}'
+}
+
+function validate_frontend() {
+    cd $WORKPATH/ui/svelte
+    local conda_env_name="OPEA_e2e"
+    export PATH=${HOME}/miniconda3/bin/:$PATH
+    if conda info --envs | grep -q "$conda_env_name"; then
+        echo "$conda_env_name exist!"
+    else
+        conda create -n ${conda_env_name} python=3.12 -y
+    fi
+    source activate ${conda_env_name}
+
+    sed -i "s/localhost/$ip_address/g" playwright.config.ts
+
+    conda install -c conda-forge nodejs=22.6.0 -y
+    npm install && npm ci && npx playwright install --with-deps
+    node -v && npm -v && pip list
+
+    exit_status=0
+    npx playwright test || exit_status=$?
+
+    if [ $exit_status -ne 0 ]; then
+        echo "[TEST INFO]: ---------frontend test failed---------"
+        exit $exit_status
+    else
+        echo "[TEST INFO]: ---------frontend test passed---------"
+    fi
+}
+
+function stop_docker() {
+    cd $WORKPATH/docker_compose/amd/gpu/rocm/
+    docker compose -f compose_vllm.yaml stop && docker compose -f compose_vllm.yaml rm -f
+}
+
+function main() {
+
+    stop_docker
+
+    if [[ "$IMAGE_REPO" == "opea" ]]; then build_docker_images; fi
+    start_services
+
+    validate_microservices
+    validate_megaservice
+    validate_frontend
+
+    stop_docker
+    echo y | docker system prune
+
+}
+
+main