[mcore] moonlight (small model with deepseekv3 arch) (#1284) #14
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: sgl | |
| on: | |
| workflow_dispatch: # Manual | |
| # Trigger the workflow on push or pull request, | |
| # but only for the main branch | |
| push: | |
| branches: | |
| - main | |
| - v0.2.x | |
| paths: | |
| - "**/*.py" | |
| - .github/workflows/vllm.yml | |
| pull_request: | |
| branches: | |
| - main | |
| - v0.2.x | |
| paths: | |
| - "**/*.py" | |
| # Other entrypoints | |
| - "!examples/**" | |
| - "!tests/**" | |
| - "!verl/trainer/main_*.py" | |
| - "!verl/trainer/fsdp_sft_trainer.py" | |
| # FSDP | |
| - "!verl/workers/**/*dp_*.py" | |
| # Megatron | |
| - "!verl/workers/**/megatron_*.py" | |
| # vLLM | |
| - "!**/*vllm*" | |
| # Recipes | |
| - "!recipe/**" | |
| # Entrypoints | |
| - ".github/workflows/sgl.yml" | |
| - "tests/rollout/*sglang*" | |
| - "tests/rollout/async_rollout_utils.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' }} | |
| # Declare permissions just read content. | |
| permissions: | |
| contents: read | |
| jobs: | |
| sgl: | |
| runs-on: [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: 1 | |
| SGL_DISABLE_TP_MEMORY_INBALANCE_CHECK: "True" | |
| container: | |
| image: ocss884/verl-sglang:ngc-th2.6.0-cu126-sglang0.4.6.post4 | |
| options: --gpus all --shm-size=10g | |
| steps: | |
| - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 | |
| with: | |
| fetch-depth: 0 | |
| - name: Install the current repository | |
| run: | | |
| pip3 install hf_transfer | |
| pip3 install -e .[test,gpu,sglang] --no-deps | |
| - name: Download Model to Use | |
| run: | | |
| huggingface-cli download 'Qwen/Qwen2-7B-Instruct' | |
| export HF_HUB_OFFLINE=1 | |
| - name: Test the latest SGLang | |
| run: | | |
| cd tests/workers/rollout | |
| torchrun --nnodes=1 --nproc_per_node=4 $(which pytest) -s test_sglang_spmd.py | |
| - name: Test the latest SGLang async | |
| run: | | |
| cd tests/workers/rollout | |
| torchrun --nnodes=1 --nproc_per_node=2 $(which pytest) -s test_sglang_async_spmd.py | |
| - name: Test the latest SGLang Rollout async with tool | |
| run: | | |
| cd tests/workers/rollout | |
| torchrun --nnodes=1 --nproc_per_node=2 $(which pytest) -s test_sglang_async_rollout_w_tools.py |