test_sdxl.yml

name: SDXL E2E Pipeline CI

on:
  workflow_dispatch:
  pull_request:
  schedule:
    - cron: "*/50 * * * *"

concurrency:
  # A PR number if a pull request and otherwise the commit hash. This cancels
  # queued and in-progress runs for the same PR (presubmit) or commit
  # (postsubmit). The workflow name is prepended to avoid conflicts between
  # different workflows.
  group: ${{ github.workflow }}-${{ github.event.number || github.sha }}
  cancel-in-progress: true

jobs:
  test-sdxl-models:
    strategy:
      matrix:
        version: [3.11]
        os: [nodai-amdgpu-mi300-x86-64]

    runs-on: ${{matrix.os}}
    env:
      IREE_TOKEN: ${{ secrets.IREE_TOKEN }}
    steps:
      - name: "Setting up Python"
        uses: actions/setup-python@75f3110429a8c05be0e1bf360334e4cced2b63fa # v2.3.3
        with:
          python-version: ${{matrix.version}}

      - name: "Checkout SHARK-ModelDev"
        uses: actions/checkout@v4
        with:
          ref: bump-punet-tom

      - name: "Checkout iree-turbine"
        uses: actions/checkout@v4
        with:
          repository: iree-org/iree-turbine
          path: iree-turbine
      
      - name: "Checkout iree"
        uses: actions/checkout@v4
        with:
          repository: iree-org/iree
          path: iree

      - name: Python deps
        run: |
          python3.11 -m venv sdxl_venv
          source sdxl_venv/bin/activate
          python -m pip install --upgrade pip
          pip install --no-compile -r ${{ github.workspace }}/iree-turbine/pytorch-cpu-requirements.txt
          pip install --pre --upgrade -r ${{ github.workspace }}/iree-turbine/requirements.txt
          pip install --no-compile --pre --upgrade -e models -r models/requirements.txt
          pip uninstall torch torchvision torchaudio -y
          pip install https://download.pytorch.org/whl/nightly/pytorch_triton_rocm-3.0.0%2B21eae954ef-cp311-cp311-linux_x86_64.whl
          pip install https://download.pytorch.org/whl/nightly/rocm6.1/torch-2.5.0.dev20240710%2Brocm6.1-cp311-cp311-linux_x86_64.whl
          pip install https://download.pytorch.org/whl/nightly/rocm6.1/torchvision-0.20.0.dev20240711%2Brocm6.1-cp311-cp311-linux_x86_64.whl
          pip install https://download.pytorch.org/whl/nightly/rocm6.1/torchaudio-2.4.0.dev20240711%2Brocm6.1-cp311-cp311-linux_x86_64.whl
          pip uninstall iree-compiler iree-runtime iree-base-compiler iree-base-runtime -y
          python ci-tools/latest-pkgci.py
          cd wheels
          unzip *.zip
          pip install *.whl
          cd ..
          rm -rf wheels

      - name: Show current free memory
        run: |
          free -mh

      - name: Run sdxl tests
        run: |
          source sdxl_venv/bin/activate
          python3 models/turbine_models/custom_models/sd_inference/sd_pipeline.py --device=hip --precision=fp16 --iree_target_triple=gfx942 --external_weights=safetensors --hf_model_name=stabilityai/stable-diffusion-xl-base-1.0 --width=1024 --height=1024 --batch_size=1 --use_i8_punet --attn_spec=punet --vae_decomp_attn --external_weights=safetensors --num_inference_steps=20 --benchmark=all --verbose