[CI] Add pytest markers to current tests and update the doc. (#577)

Signed-off-by: Alicia <115451386+congw729@users.noreply.github.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
AliciaandHongsheng Liu authored and GitHub committed 2026-02-09 16:49:45 +08:00
1 parent cdde401813
commit d3ea94309d
37 files changed
+165 -94

No files matched your search

@@ -15,6 +15,7 @@ from pathlib import Path
import pytest
import torch
from tests.utils import hardware_test
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
# ruff: noqa: E402
@@ -32,6 +33,10 @@ os.environ["VLLM_TEST_CLEAN_GPU_MEMORY"] = "1"
models = ["riverclouds/qwen_image_random"]
@pytest.mark.core_model
@pytest.mark.diffusion
@pytest.mark.cache
@hardware_test(res={"cuda": "L4", "rocm": "MI325"})
@pytest.mark.parametrize("model_name", models)
def test_cache_dit(model_name: str):
"""Test cache-dit backend with diffusion model."""
@@ -5,7 +5,7 @@ import pytest
import torch
from vllm.distributed.parallel_state import cleanup_dist_env_and_memory
from tests.utils import GPUMemoryMonitor
from tests.utils import GPUMemoryMonitor, hardware_test
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
from vllm_omni.platforms import current_omni_platform
@@ -45,6 +45,9 @@ def inference(model_name: str, offload: bool = True):
return peak
@pytest.mark.core_model
@pytest.mark.diffusion
@hardware_test(res={"cuda": "L4", "rocm": "MI325"})
@pytest.mark.skipif(current_omni_platform.is_npu() or current_omni_platform.is_rocm(), reason="Hardware not supported")
@pytest.mark.parametrize("model_name", models)
def test_cpu_offload_diffusion_model(model_name: str):
@@ -13,10 +13,10 @@ from vllm.assets.video import VideoAsset
from vllm.envs import VLLM_USE_MODELSCOPE
from vllm.multimodal.image import convert_image_mode
from tests.utils import create_new_process_for_each_test, hardware_test
from vllm_omni.platforms import current_omni_platform
from .conftest import OmniRunner
from .utils import create_new_process_for_each_test
models = ["Qwen/Qwen2.5-Omni-3B"]
@@ -34,8 +34,10 @@ test_params = [(model, stage_config) for model in models]
@pytest.mark.core_model
@pytest.mark.parametrize("test_config", test_params)
@pytest.mark.omni
@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards={"cuda": 4, "rocm": 2})
@create_new_process_for_each_test("spawn")
@pytest.mark.parametrize("test_config", test_params)
def test_mixed_modalities_to_audio(omni_runner: type[OmniRunner], test_config: tuple[str, str]) -> None:
"""Test processing audio, image, and video together, generating audio output."""
model, stage_config_path = test_config
@@ -94,8 +96,10 @@ def test_mixed_modalities_to_audio(omni_runner: type[OmniRunner], test_config: t
@pytest.mark.core_model
@pytest.mark.parametrize("test_config", test_params)
@pytest.mark.omni
@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards={"cuda": 4, "rocm": 2})
@create_new_process_for_each_test("spawn")
@pytest.mark.parametrize("test_config", test_params)
def test_mixed_modalities_to_text_only(omni_runner: type[OmniRunner], test_config: tuple[str, str]) -> None:
"""Test processing audio, image, and video together, generating audio output."""
model, stage_config_path = test_config
@@ -14,6 +14,7 @@ from pathlib import Path
import pytest
from vllm.assets.video import VideoAsset
from tests.utils import hardware_test
from vllm_omni.platforms import current_omni_platform
from .conftest import OmniRunner
@@ -31,6 +32,9 @@ else:
test_params = [(model, stage_config) for model in models for stage_config in stage_configs]
@pytest.mark.core_model
@pytest.mark.omni
@hardware_test(res={"cuda": "H100", "rocm": "MI325"}, num_cards=2)
@pytest.mark.parametrize("test_config", test_params)
def test_video_to_audio(omni_runner: type[OmniRunner], test_config) -> None:
"""Test processing video, generating audio output."""
@@ -20,16 +20,17 @@ import torch
import torch.distributed as dist
from PIL import Image
from tests.utils import hardware_test
from vllm_omni import Omni
from vllm_omni.diffusion.data import DiffusionParallelConfig
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
from vllm_omni.platforms import current_omni_platform
# ruff: noqa: E402
REPO_ROOT = Path(__file__).resolve().parents[3]
if str(REPO_ROOT) not in sys.path:
sys.path.insert(0, str(REPO_ROOT))
from vllm_omni import Omni
from vllm_omni.diffusion.data import DiffusionParallelConfig
from vllm_omni.platforms import current_omni_platform
# Test configuration
MODELS = ["riverclouds/qwen_image_random"]
@@ -145,9 +146,11 @@ def _run_inference(
# - warmup: whether to run warmup for this SP config
# - is_perf_test: whether this is a performance test (show speedup metrics)
SP_CONFIGS = [
(2, 1, DEFAULT_HEIGHT, DEFAULT_WIDTH, True, True), # Ulysses-2 - performance test
# Ulysses-2 - performance test
(2, 1, DEFAULT_HEIGHT, DEFAULT_WIDTH, True, True),
(1, 2, DEFAULT_HEIGHT, DEFAULT_WIDTH, True, True), # Ring-2 - performance test
(2, 2, DEFAULT_HEIGHT, DEFAULT_WIDTH, False, False), # Hybrid - correctness only
# Hybrid - correctness only
(2, 2, DEFAULT_HEIGHT, DEFAULT_WIDTH, False, False),
(4, 1, 272, 272, False, False), # Ulysses-4 - shape and correctness
]
@@ -162,6 +165,10 @@ def _get_sp_mode(ulysses_degree: int, ring_degree: int) -> str:
return f"hybrid-{ulysses_degree}x{ring_degree}"
@pytest.mark.core_model
@pytest.mark.diffusion
@pytest.mark.parallel
@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards={"cuda": 4, "rocm": 2})
@pytest.mark.parametrize("model_name", MODELS)
def test_sp_correctness(model_name: str):
"""Test that SP inference produces correct outputs and measure performance.
@@ -5,6 +5,7 @@ import numpy as np
import pytest
import torch
from tests.utils import hardware_test
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
from vllm_omni.outputs import OmniRequestOutput
@@ -19,6 +20,9 @@ from vllm_omni import Omni
models = ["linyueqian/stable_audio_random"]
@pytest.mark.core_model
@pytest.mark.diffusion
@hardware_test(res={"cuda": "L4"})
@pytest.mark.parametrize("model_name", models)
def test_stable_audio_model(model_name: str):
m = Omni(model=model_name)
@@ -5,6 +5,7 @@ from pathlib import Path
import pytest
import torch
from tests.utils import hardware_test
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
from vllm_omni.outputs import OmniRequestOutput
from vllm_omni.platforms import current_omni_platform
@@ -32,6 +33,9 @@ elif current_omni_platform.is_rocm():
models = ["Tongyi-MAI/Z-Image-Turbo"]
@pytest.mark.core_model
@pytest.mark.diffusion
@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards={"cuda": 1, "rocm": 2})
@pytest.mark.parametrize("model_name", models)
def test_diffusion_model(model_name: str):
m = None
@@ -15,6 +15,7 @@ from pathlib import Path
import pytest
import torch
from tests.utils import hardware_test
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
# ruff: noqa: E402
@@ -31,6 +32,10 @@ os.environ["VLLM_TEST_CLEAN_GPU_MEMORY"] = "1"
models = ["riverclouds/qwen_image_random"]
@pytest.mark.core_model
@pytest.mark.diffusion
@pytest.mark.cache
@hardware_test(res={"cuda": "L4", "rocm": "MI325"})
@pytest.mark.parametrize("model_name", models)
def test_teacache(model_name: str):
"""Test TeaCache backend with diffusion model."""
@@ -12,18 +12,18 @@ import torch
from PIL import Image
from vllm.distributed.parallel_state import cleanup_dist_env_and_memory
from tests.utils import GPUMemoryMonitor, hardware_test
from vllm_omni import Omni
from vllm_omni.diffusion.data import DiffusionParallelConfig
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
from vllm_omni.outputs import OmniRequestOutput
from vllm_omni.platforms import current_omni_platform
# ruff: noqa: E402
REPO_ROOT = Path(__file__).resolve().parents[2]
if str(REPO_ROOT) not in sys.path:
sys.path.insert(0, str(REPO_ROOT))
from tests.utils import GPUMemoryMonitor
from vllm_omni import Omni
from vllm_omni.diffusion.data import DiffusionParallelConfig
from vllm_omni.outputs import OmniRequestOutput
from vllm_omni.platforms import current_omni_platform
# os.environ["VLLM_TEST_CLEAN_GPU_MEMORY"] = "1"
os.environ["VLLM_WORKER_MULTIPROC_METHOD"] = "spawn"
@@ -127,7 +127,10 @@ def _run_zimage_generate(
cleanup_dist_env_and_memory()
@pytest.mark.integration
@pytest.mark.core_model
@pytest.mark.diffusion
@pytest.mark.parallel
@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards={"cuda": 4, "rocm": 2})
def test_zimage_tensor_parallel_tp2(tmp_path: Path):
if current_omni_platform.is_npu() or current_omni_platform.is_rocm():
pytest.skip("Z-Image TP e2e test is only supported on CUDA for now.")