LM/vLLM

compose

services:
  vllm-xpu:
    image: vllm/vllm-openai-xpu:latest
    container_name: vllm-xpu-qwen38
    privileged: true
    network_mode: host
    shm_size: "32g"

    devices:
      - /dev/dri:/dev/dri

    volumes:
      - ${HOME}/.cache/huggingface:/root/.cache/huggingface

    environment:
      - HF_HOME=/root/.cache/huggingface
      - HF_TOKEN=${HF_TOKEN:-}
      - no_proxy=localhost,127.0.0.1
      - ZE_FLAT_DEVICE_HIERARCHY=COMPOSITE
      - ZE_AFFINITY_MASK=0
      - VLLM_WORKER_MULTIPROC_METHOD=spawn
      - PYTORCH_ALLOC_CONF=expandable_segments:True
      - VLLM_XPU_ENABLE_XPU_GRAPH=1

    ports:
      - "9000:9000"

    entrypoint: ["vllm", "serve"]
    command:
      - "SergiioB/Qwen3.8-27B-GPTQ-Int4-sym-G128-MTP-BF16"
      - "--served-model-name=qwen38"
      - "--port=9000"
      - "--host=0.0.0.0"
      - "--quantization=gptq"
      - "--dtype=float16"
      - "--max-model-len=131072"
      - "--gpu-memory-utilization=0.88"
      - "--kv-cache-dtype=fp8"
      - "--max-num-seqs=1"
      - "--max-num-batched-tokens=8192"
      - "--trust-remote-code"
      - "--enable-auto-tool-choice"
      - "--tool-call-parser=qwen3_xml"
      - "--reasoning-parser=qwen3"
      - "--limit-mm-per-prompt={\"image\":10}"
      - "--speculative-config={\"method\":\"mtp\",\"num_speculative_tokens\":4}"

    restart: no

sudo usermod -aG docker $USER
docker compose up

docker.io

sudo apt install docker.io
sudo docker pull intel/vllm

docker run -t -d --shm-size 10g --net=host --ipc=host --privileged \
  -v /dev/dri/by-path:/dev/dri/by-path --name=vllm-test \
  --device /dev/dri:/dev/dri --entrypoint=/bin/bash intel/vllm:latest

source

uv venv --python 3.12 --seed --managed-python
source .venv/bin/activate

git clone https://github.com/vllm-project/vllm.git
cd vllm
pip install --upgrade pip
pip install -v -r requirements/xpu.txt

pip uninstall -y triton triton-xpu
pip install triton-xpu==3.7.2 --extra-index-url https://download.pytorch.org/whl/xpu

VLLM_TARGET_DEVICE=xpu pip install --no-build-isolation -e . -v

# check device
python -c "import torch; print(torch.xpu.is_available()); print(torch.xpu.device_count())"

# download model
hf download ciocan/gemma-4-E4B-it-W4A16

# Run
vllm serve unsloth/gemma-4-E4B-it-qat-GGUF --device xpu

Path

TODO

• https://hub.docker.com/r/intel/vllm
• https://hub.docker.com/u/intel/?page=1&search=vllm
• https://docs.vllm.ai/en/stable/getting_started/installation/gpu/#requirements
• https://www.pugetsystems.com/labs/articles/intel-arc-pro-b70-multi-gpu-ai-inference-performance/
• https://github.com/Hal9000AIML/arc-pro-b70-inference-setup-ubuntu-server
• https://sergiiob.dev/posts/intel-arc-b70-vllm-initial-mxfp4-test/
• https://forum.level1techs.com/t/intel-b70-launch-unboxed-and-tested/247873
• https://huggingface.co/blog/MatrixYao/intel-gpu