steps:
  - label: "Build wheel - CUDA 12.1"
    agents:
      queue: cpu_queue
    commands:
      - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=12.1.0 --tag vllm-ci:build-image --target build --progress plain ."
      - "mkdir artifacts"
      - "docker run --rm -v $(pwd)/artifacts:/artifacts_host vllm-ci:build-image bash -c 'cp -r dist /artifacts_host && chmod -R a+rw /artifacts_host'"
      - "bash .buildkite/upload-wheels.sh"
    env:
      DOCKER_BUILDKIT: "1"

  # Note(simon): We can always build CUDA 11.8 wheel to ensure the build is working.
  # However, this block can be uncommented to save some compute hours.
  # - block: "Build CUDA 11.8 wheel"
  #   key: block-build-cu118-wheel

  - label: "Build wheel - CUDA 11.8"
    # depends_on: block-build-cu118-wheel
    agents:
      queue: cpu_queue
    commands:
      - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg GIT_REPO_CHECK=1 --build-arg CUDA_VERSION=11.8.0 --tag vllm-ci:build-image --target build --progress plain ."
      - "mkdir artifacts"
      - "docker run --rm -v $(pwd)/artifacts:/artifacts_host vllm-ci:build-image bash -c 'cp -r dist /artifacts_host && chmod -R a+rw /artifacts_host'"
      - "bash .buildkite/upload-wheels.sh"
    env:
      DOCKER_BUILDKIT: "1"