verl/.github/workflows/model.yml

# # Tests layout

# Each folder under tests/ corresponds to a test category for a sub-namespace in verl. For instance:
# - `tests/trainer` for testing functionality related to `verl/trainer`
# - `tests/models` for testing functionality related to `verl/models`
# - ...

# There are a few folders with `special_` prefix, created for special purposes:
# - `special_distributed`: unit tests that must run with multiple GPUs
# - `special_e2e`: end-to-end tests with training/generation scripts
# - `special_npu`: tests for NPUs
# - `special_sanity`: a suite of quick sanity tests
# - `special_standalone`: a set of test that are designed to run in dedicated environments

# Accelerators for tests 
# - By default tests are run with GPU available, except for the ones under `special_npu`, and any test script whose name ends with `on_cpu.py`.
# - For test scripts with `on_cpu.py` name suffix would be tested on CPU resources in linux environment.

# # Workflow layout

# All CI tests are configured by yaml files in `.github/workflows/`. Here's an overview of all test configs:
# 1. A list of always triggered CPU sanity tests: `check-pr-title.yml`, `secrets_scan.yml`, `check-pr-title,yml`, `pre-commit.yml`, `doc.yml`
# 2. Some heavy multi-GPU unit tests, such as `model.yml`, `vllm.yml`, `sgl.yml`
# 3. End-to-end tests: `e2e_*.yml`
# 4. Unit tests
#   - `cpu_unit_tests.yml`, run pytest on all scripts with file name pattern `tests/**/test_*_on_cpu.py`
#   - `gpu_unit_tests.yml`, run pytest on all scripts with file without the `on_cpu.py` suffix.
#   - Since cpu/gpu unit tests by default runs all tests under `tests`, please make sure tests are manually excluded in them when
#     - new workflow yaml is added to `.github/workflows`
#     - new tests are added to workflow mentioned in 2.
# name: Check PR Title

name: model

on:
  # Trigger the workflow on push or pull request,
  # but only for the main branch
  push:
    branches:
      - main
      - v0.*
  pull_request:
    branches:
      - main
      - v0.*
    paths:
      - "verl/**/*.py"
      # Entrypoints
      - ".github/workflows/model.yml"
      - "tests/special_distributed/test_fsdp_ckpt.py"
      - "tests/special_distributed/test_mcore_config_converter.py"
      - "tests/special_distributed/test_tensor_dict.py"
      - "tests/models/**"
      - "tests/special_distributed/run_all.sh"

# Declare permissions just read content.
permissions:
  contents: read

# Cancel jobs on the same ref if a new one is triggered
concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}


env:
  IMAGE: "verl-ci-cn-beijing.cr.volces.com/verlai/verl:app-verl0.5-transformers4.55.4-vllm0.10.0-mcore0.13.0-te2.2"
  DYNAMIC_RUNNER_ENDPOINT: "https://sd10g3clalm04ug7alq90.apigateway-cn-beijing.volceapi.com/runner"

jobs:
  setup:
    if: github.repository_owner == 'volcengine'
    runs-on: ubuntu-latest
    outputs:
      runner-label: ${{ steps.create-runner.outputs.runner-label }}
      mlp-task-id: ${{ steps.create-runner.outputs.mlp-task-id }}
    steps:
      - uses: actions/checkout@v4
      - id: create-runner
        uses: volcengine/vemlp-github-runner@v1
        with:
          mode: "create"
          faas-url: "${{ env.DYNAMIC_RUNNER_ENDPOINT }}"
          mlp-image: "${{ env.IMAGE }}"

  model_rmpad:
    needs: setup
    runs-on: ["${{ needs.setup.outputs.runner-label || 'L20x8' }}"]
    timeout-minutes: 20 # Increase this timeout value as needed
    env:
      HTTP_PROXY: ${{ secrets.PROXY_HTTP }}
      HTTPS_PROXY: ${{ secrets.PROXY_HTTPS }}
      NO_PROXY: "localhost,127.0.0.1,hf-mirror.com"
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
        with:
          fetch-depth: 0
      - name: Install the current repository and upgrade to latest transformers(4.54.0)/flash_attn, transformers 4.55.0 has strange behavior with model backward
        run: |
          pip3 install --no-deps -e .[test]
          pip3 install --upgrade transformers
      - name: Running rmpad model tests on 8 L20 GPUs + flash_attn 2.5.8
        run: |
          pytest -s tests/models/test_transformer.py
      - name: Running rmpad model tests on 8 L20 GPUs + latest flash_attn
        run: |
          pytest -s tests/models/test_transformer.py
      - name: Running FSDP rmpad model tests on 8 L20 GPUs + latest flash_attn
        run: |
          STRATEGY=fsdp torchrun --nproc_per_node=8 tests/special_distributed/test_fsdp_ckpt.py
      - name: Running transformers ulysses tests on 8 L20 GPUs + latest transformers
        run: |
          torchrun --nproc_per_node=8 -m pytest tests/models/test_transformers_ulysses.py
      - name: Running transformers ulysses tests on 8 L20 GPUs + transformers 4.54.1
        run: |
          pip3 install transformers==4.54.1
          torchrun --nproc_per_node=8 -m pytest tests/models/test_transformers_ulysses.py
      - name: Running transformers ulysses tests on 8 L20 GPUs + transformers 4.53.2
        run: |
          pip3 install transformers==4.53.2
          torchrun --nproc_per_node=8 -m pytest tests/models/test_transformers_ulysses.py
      - name: Running transformers ulysses tests on 8 L20 GPUs + transformers 4.52.0
        run: |
          pip3 install transformers==4.52.0
          torchrun --nproc_per_node=8 -m pytest tests/models/test_transformers_ulysses.py
      - name: Run distributed test
        run: |
          bash tests/special_distributed/run_all.sh

  # TODO: Move this back to model_rmpad once FSDP2 is stable.
  # NOTE: List as an independent job to make rerun easier.
  model_rmpad_fsdp2_unstable:
    needs: setup
    runs-on: [ "${{ needs.setup.outputs.runner-label || 'L20x8' }}" ]
    timeout-minutes: 20 # Increase this timeout value as needed
    env:
      HTTP_PROXY: ${{ secrets.PROXY_HTTP }}
      HTTPS_PROXY: ${{ secrets.PROXY_HTTPS }}
      NO_PROXY: "localhost,127.0.0.1,hf-mirror.com"
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
        with:
          fetch-depth: 0
      - name: Install the current repository and upgrade to latest transformers/flash_attn
        run: |
          pip3 install --no-deps -e .[test]
          pip3 install --upgrade transformers
      - name: Running FSDP2 rmpad model tests on 8 L20 GPUs + latest flash_attn
        run: |
          STRATEGY=fsdp2 torchrun --nproc_per_node=8 tests/special_distributed/test_fsdp_ckpt.py

  mcore_config_converter:
    needs: setup
    runs-on: [ "${{ needs.setup.outputs.runner-label || 'L20x8' }}" ]
    timeout-minutes: 20 # Increase this timeout value as needed
    env:
      HTTP_PROXY: ${{ secrets.PROXY_HTTP }}
      HTTPS_PROXY: ${{ secrets.PROXY_HTTPS }}
      NO_PROXY: "localhost,127.0.0.1,hf-mirror.com"
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
        with:
          fetch-depth: 0
      - name: Install the current repository
        run: |
          pip3 install --no-deps -e .[test]
          pip install --upgrade "huggingface_hub[cli]"
#      - name: Download model config files
#        run: |
#          hf download Qwen/Qwen2.5-7B config.json --local-dir $HOME/configs/Qwen/Qwen2.5-7B
#          hf download Qwen/Qwen3-8B config.json --local-dir $HOME/configs/Qwen/Qwen3-8B
#          hf download deepseek-ai/deepseek-coder-1.3b-instruct config.json --local-dir $HOME/configs/deepseek-ai/deepseek-coder-1.3b-instruct
#          hf download Qwen/Qwen2-57B-A14B config.json --local-dir $HOME/configs/Qwen/Qwen2-57B-A14B
#          hf download Qwen/Qwen3-30B-A3B config.json --local-dir $HOME/configs/Qwen/Qwen3-30B-A3B
#          hf download deepseek-ai/DeepSeek-V3-Base config.json --local-dir $HOME/configs/deepseek-ai/DeepSeek-V3-Base
      - name: Running mcore config converter tests on 8 L20 GPUs
        run: |
          torchrun --nproc_per_node=8 tests/special_distributed/test_mcore_config_converter.py

  model_engine:
    needs: setup
    runs-on: [ "${{ needs.setup.outputs.runner-label || 'L20x8' }}" ]
    timeout-minutes: 20 # Increase this timeout value as needed
    env:
      HTTP_PROXY: ${{ secrets.PROXY_HTTP }}
      HTTPS_PROXY: ${{ secrets.PROXY_HTTPS }}
      NO_PROXY: "localhost,127.0.0.1,hf-mirror.com"
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
        with:
          fetch-depth: 0
      - name: Install the current repository
        run: |
          pip3 install --no-deps -e .[test]
          pip3 install --upgrade tensordict transformers
          pip install --upgrade "huggingface_hub[cli]"
      - name: Download model config files
        run: |
          hf download Qwen/Qwen2.5-0.5B-Instruct --local-dir $HOME/models/Qwen/Qwen2.5-0.5B-Instruct

      - name: Running mcore engine tests on 8 L20 GPUs
        run: |
          pytest -s -x tests/models/test_engine.py

  cleanup:
    runs-on: ubuntu-latest
    needs:
      [
        setup,
        model_rmpad,
        model_rmpad_fsdp2_unstable,
        mcore_config_converter,
        model_engine
      ]
    if: always()
    steps:
      - id: destroy-runner
        uses: volcengine/vemlp-github-runner@v1
        with:
          mode: "destroy"
          faas-url: "${{ env.DYNAMIC_RUNNER_ENDPOINT }}"
          mlp-task-id: "${{ needs.setup.outputs.mlp-task-id }}"