GenAIExamples/CodeTrans/docker_compose/amd/gpu/rocm/compose_vllm.yaml

# Copyright (C) 2024 Intel Corporation
# Copyright (c) 2024 Advanced Micro Devices, Inc.
# SPDX-License-Identifier: Apache-2.0

services:
  codetrans-vllm-service:
    image: ${REGISTRY:-opea}/vllm-rocm:${TAG:-latest}
    container_name: codetrans-vllm-service
    ports:
      - "${CODETRANS_VLLM_SERVICE_PORT:-8081}:8011"
    environment:
      no_proxy: ${no_proxy}
      http_proxy: ${http_proxy}
      https_proxy: ${https_proxy}
      HUGGINGFACEHUB_API_TOKEN: ${CODETRANS_HUGGINGFACEHUB_API_TOKEN}
      HF_TOKEN: ${CODETRANS_HUGGINGFACEHUB_API_TOKEN}
      HF_HUB_DISABLE_PROGRESS_BARS: 1
      HF_HUB_ENABLE_HF_TRANSFER: 0
      WILM_USE_TRITON_FLASH_ATTENTION: 0
      PYTORCH_JIT: 0
    healthcheck:
      test: [ "CMD-SHELL", "curl -f http://${HOST_IP}:${CODETRANS_VLLM_SERVICE_PORT:-8028}/health || exit 1" ]
      interval: 10s
      timeout: 10s
      retries: 100
    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 ${CODETRANS_LLM_MODEL_ID} --swap-space 16 --disable-log-requests --dtype float16 --tensor-parallel-size 4 --host 0.0.0.0 --port 8011 --num-scheduler-steps 1 --distributed-executor-backend \"mp\""
    ipc: host
  codetrans-llm-server:
    image: ${REGISTRY:-opea}/llm-textgen:${TAG:-latest}
    container_name: codetrans-llm-server
    depends_on:
      codetrans-vllm-service:
        condition: service_healthy
    ports:
      - "${CODETRANS_LLM_SERVICE_PORT:-9000}:9000"
    ipc: host
    environment:
      no_proxy: ${no_proxy}
      http_proxy: ${http_proxy}
      https_proxy: ${https_proxy}
      LLM_ENDPOINT: ${CODETRANS_LLM_ENDPOINT}
      LLM_MODEL_ID: ${CODETRANS_LLM_MODEL_ID}
      HUGGINGFACEHUB_API_TOKEN: ${CODETRANS_HUGGINGFACEHUB_API_TOKEN}
      HF_TOKEN: ${CODETRANS_HUGGINGFACEHUB_API_TOKEN}
      LLM_COMPONENT_NAME: "OpeaTextGenService"
    restart: unless-stopped
  codetrans-backend-server:
    image: ${REGISTRY:-opea}/codetrans:${TAG:-latest}
    container_name: codetrans-backend-server
    depends_on:
      - codetrans-llm-server
    ports:
      - "${CODETRANS_BACKEND_SERVICE_PORT:-7777}:7777"
    environment:
      no_proxy: ${no_proxy}
      https_proxy: ${https_proxy}
      http_proxy: ${http_proxy}
      MEGA_SERVICE_HOST_IP: ${HOST_IP}
      LLM_SERVICE_HOST_IP: ${HOST_IP}
      LLM_SERVICE_PORT: ${CODETRANS_LLM_SERVICE_PORT}
    ipc: host
    restart: always
  codetrans-ui-server:
    image: ${REGISTRY:-opea}/codetrans-ui:${TAG:-latest}
    container_name: codetrans-ui-server
    depends_on:
      - codetrans-backend-server
    ports:
      - "${CODETRANS_FRONTEND_SERVICE_PORT:-5173}:5173"
    environment:
      no_proxy: ${no_proxy}
      https_proxy: ${https_proxy}
      http_proxy: ${http_proxy}
      BASE_URL: ${CODETRANS_BACKEND_SERVICE_URL}
      BASIC_URL: ${CODETRANS_BACKEND_SERVICE_URL}
    ipc: host
    restart: always
  codetrans-nginx-server:
    image: ${REGISTRY:-opea}/nginx:${TAG:-latest}
    container_name: codetrans-nginx-server
    depends_on:
      - codetrans-backend-server
      - codetrans-ui-server
    ports:
      - "${CODETRANS_NGINX_PORT:-80}:80"
    environment:
      - no_proxy=${no_proxy}
      - https_proxy=${https_proxy}
      - http_proxy=${http_proxy}
      - FRONTEND_SERVICE_IP=${CODETRANS_FRONTEND_SERVICE_IP}
      - FRONTEND_SERVICE_PORT=${CODETRANS_FRONTEND_SERVICE_PORT}
      - BACKEND_SERVICE_NAME=${CODETRANS_BACKEND_SERVICE_NAME}
      - BACKEND_SERVICE_IP=${CODETRANS_BACKEND_SERVICE_IP}
      - BACKEND_SERVICE_PORT=${CODETRANS_BACKEND_SERVICE_PORT}
    ipc: host
    restart: always

networks:
  default:
    driver: bridge