Fix CI: Install CUDA toolkit for flashinfer JIT compilation #56

	name: Integration Tests (8 card)

	on:
	push:
	branches: [ main, fix-ci-cuda-toolkit ] # TODO: remove fix-ci-cuda-toolkit after testing
	workflow_dispatch:

	concurrency:
	group: integration-test-${{ github.workflow }}-${{ github.ref == 'refs/heads/main' && github.run_number \|\| github.ref }}
	cancel-in-progress: true

	permissions:
	id-token: write
	contents: read

	defaults:
	run:
	shell: bash -l -eo pipefail {0}

	jobs:
	integration_test:
	if: github.repository_owner == 'meta-pytorch'
	runs-on: linux.g4dn.metal.nvidia.gpu
	steps:
	- name: Check out repo
	uses: actions/checkout@v4
	- name: Setup conda env
	uses: conda-incubator/setup-miniconda@v2
	with:
	auto-update-conda: true
	miniconda-version: "latest"
	activate-environment: test
	python-version: '3.12'
	- name: Update pip
	run: python -m pip install --upgrade pip
	- name: Install CUDA toolkit
	run: \|
	# flashinfer (used by vLLM 0.13.0) requires nvcc for JIT compilation
	sudo dnf install -y cuda-toolkit-12-8
	echo "CUDA_HOME=/usr/local/cuda-12.8" >> $GITHUB_ENV
	echo "/usr/local/cuda-12.8/bin" >> $GITHUB_PATH
	- name: Install torchforge
	run: pip install uv && uv pip install . && uv pip install .[dev]
	- name: Run weight sync integration test
	run: pytest -s tests/integration_tests/test_policy_update.py::TestWeightSync::test_sanity_check --config tests/integration_tests/fixtures/qwen3_1_7b_tp.yaml

Navigation Menu