e2e_ascend.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: e2e_ascend

on:
  # Trigger the workflow on push or pull request,
  # but only for the main branch
  push:
    branches:
      - main
      - v0.*
  pull_request:
    branches:
      - main
    paths:
      - ".github/workflows/e2e_ascend.yml"
      - "examples/data_preprocess/**"
      - "examples/grpo_trainer/**"
      - "examples/ppo_trainer/**"
      - "examples/sft/**"
      - "verl/experimental/one_step_off_policy/**"
      - "tests/special_npu/**"
      - "tests/special_sanity/check_device_api_usage.py"
      - "verl/**"
      - "pyproject.toml"
      - "requirements-npu.txt"
      - "setup.py"

# 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' }}

permissions:
  contents: read

jobs:
  non_rl_job:
    if: github.repository_owner == 'volcengine'
    name: E2E Ascend testing for non-RL algorithm scenarios
    runs-on: linux-aarch64-a2-4
    timeout-minutes: 60
    container:
      image: swr.ap-southeast-1.myhuaweicloud.com/base_image/ascend-ci/verl/verl:verl-8.3.rc1-910b-ubuntu22.04-py3.11-latest
      options: >-
        --shm-size 16g
    env:
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - name: Config third-party dependency download cache
        run: |
          sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
          pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
          pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
      - name: Check npu and CANN info
        run: |
          cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
          npu-smi info
      - name: Check initial pip list from image
        run: |
          pip list
      - name: Checkout volcengine/verl repo
        uses: actions/checkout@v4
        with:
          fetch-depth: 0
          clean: true
      - name: Install the current repository
        run: |
          pip install -r requirements-npu.txt
          pip install -e .
      - name: Check final pip list
        run: |
          pip list
      - name: Preprocess gsm8k dataset
        run: |
          python examples/data_preprocess/gsm8k_multiturn_sft.py --local_dataset_path ${HOME}/.cache/datasets/openai/gsm8k
      - name: Running gsm8k e2e training tests with peft sft on ASCEND NPU
        run: |
          ray stop --force
          bash tests/special_npu/run_qwen2_5_05b_sft_peft_sp2.sh
      - name: Running gsm8k e2e training tests with peft sft on ASCEND NPU (ray mode)
        run: |
          ray stop --force
          mode=ray bash tests/special_npu/run_qwen2_5_05b_sft_peft_sp2.sh

  llm_rl_job:
    if: github.repository_owner == 'volcengine'
    name: E2E Ascend testing for RL training scenarios of LLM models
    runs-on: linux-aarch64-a2-8
    timeout-minutes: 60
    container:
      image: swr.ap-southeast-1.myhuaweicloud.com/base_image/ascend-ci/verl/verl:verl-8.3.rc1-910b-ubuntu22.04-py3.11-latest
      options: >-
        --shm-size 16g
    env:
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - name: Config third-party dependency download cache
        run: |
          sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
          pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
          pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
      - name: Check npu and CANN info
        run: |
          cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
          npu-smi info
      - name: Check initial pip list from image
        run: |
          pip list
      - name: Checkout volcengine/verl repo
        uses: actions/checkout@v4
        with:
          fetch-depth: 0
          clean: true
      - name: Install the current repository
        run: |
          pip install -r requirements-npu.txt
          pip install -e .
      - name: Check final pip list
        run: |
          pip list
      - name: Preprocess gsm8k dataset
        run: |
          python examples/data_preprocess/gsm8k.py --local_dataset_path ${HOME}/.cache/datasets/openai/gsm8k
      - name: Running gsm8k e2e training tests with PPO on ASCEND NPU (FSDP backend)
        run: |
          ray stop --force
          bash tests/special_npu/run_qwen3_06b_ppo.sh
          rm -rf $HOME/ckpts
      - name: Running gsm8k e2e training tests with GRPO on ASCEND NPU (FSDP backend)
        run: |
          ray stop --force
          bash tests/special_npu/run_qwen2_5_05b_grpo.sh
          rm -rf $HOME/ckpts
      - name: Running gsm8k e2e training tests with GRPO on ASCEND NPU (MindSpeed backend)
        run: |
          ray stop --force
          export PYTHONPATH=$PYTHONPATH:/Megatron-LM
          USE_DIST_CKPT=True bash tests/special_npu/run_qwen2_5_05b_grpo_mindspeed.sh
          rm -rf $HOME/dist_ckpt/qwen2_5_05b_grpo_mindspeed
          rm -rf $HOME/ckpts
      - name: Running gsm8k e2e training tests with GRPO on ASCEND NPU (MindSpeed backend, MoE Model)
        run: |
          ray stop --force
          export PYTHONPATH=$PYTHONPATH:/Megatron-LM
          USE_DIST_CKPT=True USE_DUMMY_MODEL=True DUMMY_MODEL_CONFIG_PATH=tests/special_e2e/ppo_trainer/expert_parallel/qwen3moe_minimal.json DUMMY_MODEL_PATH=$HOME/dist_ckpt/qwen3_30b_grpo_mindspeed bash tests/special_npu/run_qwen3_30b_grpo_mindspeed.sh

  vlm_rl_job:
    if: github.repository_owner == 'volcengine'
    name: E2E Ascend testing for RL training scenarios of VLM models
    runs-on: linux-aarch64-a2-8
    timeout-minutes: 60
    container:
      image: swr.ap-southeast-1.myhuaweicloud.com/base_image/ascend-ci/verl/verl:verl-8.3.rc1-910b-ubuntu22.04-py3.11-latest
      options: >-
        --shm-size 16g
    env:
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - name: Config third-party dependency download cache
        run: |
          sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
          pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
          pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
      - name: Check npu and CANN info
        run: |
          cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
          npu-smi info
      - name: Check initial pip list from image
        run: |
          pip list
      - name: Checkout volcengine/verl repo
        uses: actions/checkout@v4
        with:
          fetch-depth: 0
          clean: true
      - name: Install the current repository
        run: |
          pip install -r requirements-npu.txt
          pip install -e .
      - name: Check final pip list
        run: |
          pip list
      - name: Preprocess geo3k dataset
        run: |
          python examples/data_preprocess/geo3k.py --local_dataset_path ${HOME}/.cache/datasets/hiyouga/geometry3k
      - name: Running geo3k e2e training tests with GRPO on ASCEND NPU
        run: |
          ray stop --force
          bash tests/special_npu/run_qwen2_5_vl_3b_npu.sh
          rm -rf $HOME/ckpts

  experimental_job:
    if: github.repository_owner == 'volcengine'
    name: E2E Ascend testing for experimental features
    runs-on: linux-aarch64-a2-4
    timeout-minutes: 60
    container:
      image: swr.ap-southeast-1.myhuaweicloud.com/base_image/ascend-ci/verl/verl:verl-8.3.rc1-910b-ubuntu22.04-py3.11-latest
      options: >-
        --shm-size 16g
    env:
      HF_ENDPOINT: "https://hf-mirror.com"
      HF_HUB_ENABLE_HF_TRANSFER: "0" # This is more stable
    steps:
      - name: Check npu and CANN info
        run: |
          cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
          npu-smi info
      - name: Check initial pip list from image
        run: |
          pip list
      - name: Checkout volcengine/verl repo
        uses: actions/checkout@v4
        with:
          fetch-depth: 0
          clean: true
      - name: Install the current repository
        run: |
          pip install -r requirements-npu.txt
          pip install -e .
      - name: Check final pip list
        run: |
          pip list
      - name: Preprocess gsm8k dataset
        run: |
          python examples/data_preprocess/gsm8k.py --local_dataset_path ${HOME}/.cache/datasets/openai/gsm8k
      - name: Running the E2E test with one_step_off_policy algorithm on ASCEND NPU (FSDP2)
        run: |
          ray stop --force
          bash tests/special_npu/run_one_step_off_policy.sh
          rm -rf $HOME/ckpts