test_cli_cuda_tensorrt_llm.yaml

name: CLI CUDA TensorRT-LLM Tests

on:
  workflow_dispatch:
  push:
    branches: [main]
  pull_request:
    types: [opened, reopened, synchronize]

concurrency:
  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
  cancel-in-progress: true

jobs:
  pull_image_and_run_cli_tensorrt_llm_tests:
    runs-on: [single-gpu, nvidia-gpu, a10, ci]

    steps:
      - name: Checkout
        uses: actions/checkout@v3

      - name: Run tests
        run: docker run
          --rm
          --gpus all
          --shm-size 64G
          --env USE_CUDA="1"
          --env PROCESS_SPECIFIC_VRAM="0"
          --volume $(pwd):/workspace/optimum-benchmark
          --workdir /workspace/optimum-benchmark
          --entrypoint /bin/bash
          huggingface/optimum-nvidia:latest
          -c "pip install -e .[testing] && pip uninstall -y nvidia-ml-py &&
          pytest -s -v -k 'cli and tensorrt_llm' -x"