run slow tests aginst vllm and transformers main #5

	name: Tests on dev branch of vllm and transformers

	on:
	# Run automatically every Saturday at 00:00 UTC
	schedule:
	- cron: "0 0 * * 6"

	# Allow manual triggering via GitHub UI
	workflow_dispatch:

	# Optional: run on pushes to main or release branches
	push:
	branches:
	- main
	- v*-release

	pull_request:
	branches:
	- main

	jobs:
	run_tests:
	name: Run tests on dev branch of vllm and transformers
	runs-on: 'aws-g4dn-2xlarge-use1-public-80'
	steps:
	- name: Install Git LFS
	run: \|
	sudo apt-get update && sudo apt-get install -y git-lfs
	git lfs install

	- name: Install Python development headers
	run: sudo apt-get update && sudo apt-get install -y python3.10-dev

	- name: Checkout repository
	uses: actions/checkout@v4
	with:
	lfs: true

	- name: Install uv
	uses: astral-sh/setup-uv@v5
	with:
	enable-cache: true

	- name: Install the project
	run: \|
	uv sync --extra dev
	VLLM_USE_PRECOMPILED=1 uv pip install --upgrade git+https://github.com/vllm-project/vllm.git@main
	uv pip install --upgrade git+https://github.com/huggingface/transformers.git@main

	- name: run nvidia-smi
	run: nvidia-smi

	- name: Pip freeze
	run: uv pip freeze

	- name: Run tests
	run: \|
	VLLM_WORKER_MULTIPROC_METHOD=spawn uv run pytest --disable-pytest-warnings --runslow tests/slow_tests/test_vllm_model.py
	uv run pytest --disable-pytest-warnings --runslow tests/slow_tests/test_accelerate_model.py
	env:
	VLLM_ATTENTION_BACKEND: "FLASHINFER"
	VLLM_USE_FLASH_ATTN: "0"

Provide feedback