Qwen2-VL

LMDeploy 支持以下 Qwen-VL 系列模型，详细信息如下表所示：

模型	大小	支持的推理引擎
Qwen-VL-Chat	-	TurboMind
Qwen2-VL	2B, 7B	PyTorch

下一章将演示如何使用LMDeploy部署一个Qwen-VL模型，以Qwen2-VL-7B-Instruct为例。

安装

请按照安装指南安装LMDeploy，并安装Qwen2-VL所需的其他包

pip install qwen_vl_utils

或者，您可以构建一个docker镜像来设置推理环境。如果您的宿主机上的CUDA版本是>=12.4，您可以运行：

git clone https://github.com/InternLM/lmdeploy.git
cd lmdeploy
docker build --build-arg CUDA_VERSION=cu12 -t openmmlab/lmdeploy:qwen2vl . -f ./docker/Qwen2VL_Dockerfile

否则，你可以选择：

docker build --build-arg CUDA_VERSION=cu11 -t openmmlab/lmdeploy:qwen2vl . -f ./docker/Qwen2VL_Dockerfile

离线推理

以下示例代码展示了VLM管道的基本用法。有关详细信息，请参阅VLM离线推理管道

from lmdeploy import pipeline
from lmdeploy.vl import load_image

pipe = pipeline('Qwen/Qwen2-VL-2B-Instruct')

image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/tests/data/tiger.jpeg')
response = pipe((f'describe this image', image))
print(response)

更多示例如下：

multi-image multi-round conversation, combined images

from lmdeploy import pipeline, GenerationConfig

pipe = pipeline('Qwen/Qwen2-VL-2B-Instruct', log_level='INFO')
messages = [
    dict(role='user', content=[
        dict(type='text', text='Describe the two images in detail.'),
        dict(type='image_url', image_url=dict(url='https://raw.githubusercontent.com/QwenLM/Qwen-VL/master/assets/mm_tutorial/Beijing_Small.jpeg')),
        dict(type='image_url', image_url=dict(url='https://raw.githubusercontent.com/QwenLM/Qwen-VL/master/assets/mm_tutorial/Chongqing_Small.jpeg'))
    ])
]
out = pipe(messages, gen_config=GenerationConfig(top_k=1))

messages.append(dict(role='assistant', content=out.text))
messages.append(dict(role='user', content='What are the similarities and differences between these two images.'))
out = pipe(messages, gen_config=GenerationConfig(top_k=1))

image resolution for performance boost

from lmdeploy import pipeline, GenerationConfig

pipe = pipeline('Qwen/Qwen2-VL-2B-Instruct', log_level='INFO')

min_pixels = 64 * 28 * 28
max_pixels = 64 * 28 * 28
messages = [
    dict(role='user', content=[
        dict(type='text', text='Describe the two images in detail.'),
        dict(type='image_url', image_url=dict(min_pixels=min_pixels, max_pixels=max_pixels, url='https://raw.githubusercontent.com/QwenLM/Qwen-VL/master/assets/mm_tutorial/Beijing_Small.jpeg')),
        dict(type='image_url', image_url=dict(min_pixels=min_pixels, max_pixels=max_pixels, url='https://raw.githubusercontent.com/QwenLM/Qwen-VL/master/assets/mm_tutorial/Chongqing_Small.jpeg'))
    ])
]
out = pipe(messages, gen_config=GenerationConfig(top_k=1))

messages.append(dict(role='assistant', content=out.text))
messages.append(dict(role='user', content='What are the similarities and differences between these two images.'))
out = pipe(messages, gen_config=GenerationConfig(top_k=1))

在线服务

你可以通过lmdeploy serve api_server CLI启动服务器：

lmdeploy serve api_server Qwen/Qwen2-VL-2B-Instruct

你也可以使用上述构建的docker镜像来启动服务：

docker run --runtime nvidia --gpus all \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HUGGING_FACE_HUB_TOKEN=<secret>" \
    -p 23333:23333 \
    --ipc=host \
    openmmlab/lmdeploy:qwen2vl \
    lmdeploy serve api_server Qwen/Qwen2-VL-2B-Instruct

Docker compose 是另一种选择。在 lmdeploy 项目的根目录下创建一个 docker-compose.yml 配置文件，如下所示：

version: '3.5'

services:
  lmdeploy:
    container_name: lmdeploy
    image: openmmlab/lmdeploy:qwen2vl
    ports:
      - "23333:23333"
    environment:
      HUGGING_FACE_HUB_TOKEN: <secret>
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface
    stdin_open: true
    tty: true
    ipc: host
    command: lmdeploy serve api_server Qwen/Qwen2-VL-2B-Instruct
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: "all"
              capabilities: [gpu]

然后，您可以执行如下启动命令：

docker-compose up -d

如果你在运行docker logs -f lmdeploy后看到以下日志，这意味着服务已成功启动。

HINT:    Please open  http://0.0.0.0:23333   in a browser for detailed api usage!!!
HINT:    Please open  http://0.0.0.0:23333   in a browser for detailed api usage!!!
HINT:    Please open  http://0.0.0.0:23333   in a browser for detailed api usage!!!
INFO:     Started server process [2439]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on  http://0.0.0.0:23333  (Press CTRL+C to quit)

lmdeploy serve api_server 的参数可以通过 lmdeploy serve api_server -h 详细查看。

有关api_server的更多信息以及如何访问该服务，可以从这里找到