[model] support FLUX.1-Kontext-dev (#561)
Signed-off-by: Lancer <maruxiiang6688@gmail.com> Signed-off-by: Lancer <maruixiang6688@gmail.com> Signed-off-by: Lancer <402430575@qq.com> Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com> Signed-off-by: Gao Han <hgaoaf@connect.ust.hk> Co-authored-by: Lancer <maruxiiang6688@gmail.com> Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com> Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
This commit is contained in:
11 files changed
+1255
-126
No files matched your search
@@ -0,0 +1,126 @@
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
||||
|
||||
"""
|
||||
Tests for FluxKontext model pipeline.
|
||||
|
||||
FluxKontext is a text-to-image and image-to-image diffusion model that supports:
|
||||
- Text-to-image generation
|
||||
- Image editing with text guidance
|
||||
"""
|
||||
|
||||
import os
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
import pytest
|
||||
from PIL import Image
|
||||
|
||||
from vllm_omni.diffusion.data import DiffusionParallelConfig
|
||||
from vllm_omni.entrypoints.omni import Omni
|
||||
from vllm_omni.inputs.data import OmniDiffusionSamplingParams
|
||||
|
||||
REPO_ROOT = Path(__file__).resolve().parents[2]
|
||||
if str(REPO_ROOT) not in sys.path:
|
||||
sys.path.insert(0, str(REPO_ROOT))
|
||||
|
||||
os.environ["VLLM_TEST_CLEAN_GPU_MEMORY"] = "1"
|
||||
|
||||
MODEL = "black-forest-labs/FLUX.1-Kontext-dev"
|
||||
|
||||
|
||||
@pytest.mark.core_model
|
||||
@pytest.mark.diffusion
|
||||
def test_flux_kontext_text_to_image():
|
||||
"""Test FluxKontext text-to-image generation with real model."""
|
||||
omni = Omni(
|
||||
model=MODEL,
|
||||
parallel_config=DiffusionParallelConfig(
|
||||
tensor_parallel_size=2,
|
||||
),
|
||||
enable_cpu_offload=False,
|
||||
)
|
||||
|
||||
try:
|
||||
omni_outputs = list(
|
||||
omni.generate(
|
||||
prompts=["A photo of a cat sitting on a laptop"],
|
||||
sampling_params_list=OmniDiffusionSamplingParams(
|
||||
height=512,
|
||||
width=512,
|
||||
num_inference_steps=2,
|
||||
seed=42,
|
||||
),
|
||||
)
|
||||
)
|
||||
|
||||
assert len(omni_outputs) > 0
|
||||
output = omni_outputs[0]
|
||||
images = None
|
||||
if output.images:
|
||||
images = output.images
|
||||
elif hasattr(output, "request_output") and output.request_output:
|
||||
for stage_out in output.request_output:
|
||||
if hasattr(stage_out, "images") and stage_out.images:
|
||||
images = stage_out.images
|
||||
break
|
||||
|
||||
assert images is not None
|
||||
assert len(images) > 0
|
||||
assert isinstance(images[0], Image.Image)
|
||||
assert images[0].size == (512, 512)
|
||||
finally:
|
||||
omni.close()
|
||||
|
||||
|
||||
@pytest.mark.core_model
|
||||
@pytest.mark.diffusion
|
||||
def test_flux_kontext_image_edit():
|
||||
"""Test FluxKontext image-to-image editing with real model."""
|
||||
from vllm.assets.image import ImageAsset
|
||||
|
||||
input_image = ImageAsset("2560px-Gfp-wisconsin-madison-the-nature-boardwalk").pil_image.convert("RGB")
|
||||
omni = Omni(
|
||||
model=MODEL,
|
||||
parallel_config=DiffusionParallelConfig(
|
||||
tensor_parallel_size=2,
|
||||
),
|
||||
enable_cpu_offload=False,
|
||||
)
|
||||
|
||||
try:
|
||||
omni_outputs = list(
|
||||
omni.generate(
|
||||
prompts=[
|
||||
{
|
||||
"prompt": "Transform this image into a Vincent van Gogh style painting",
|
||||
"multi_modal_data": {"img2img": input_image},
|
||||
"modalities": ["img2img"],
|
||||
}
|
||||
],
|
||||
sampling_params_list=OmniDiffusionSamplingParams(
|
||||
height=512,
|
||||
width=512,
|
||||
num_inference_steps=2,
|
||||
seed=42,
|
||||
),
|
||||
)
|
||||
)
|
||||
|
||||
assert len(omni_outputs) > 0
|
||||
output = omni_outputs[0]
|
||||
images = None
|
||||
if output.images:
|
||||
images = output.images
|
||||
elif hasattr(output, "request_output") and output.request_output:
|
||||
for stage_out in output.request_output:
|
||||
if hasattr(stage_out, "images") and stage_out.images:
|
||||
images = stage_out.images
|
||||
break
|
||||
|
||||
assert images is not None
|
||||
assert len(images) > 0
|
||||
assert isinstance(images[0], Image.Image)
|
||||
assert images[0].size == (512, 512)
|
||||
finally:
|
||||
omni.close()
|
||||
@@ -0,0 +1,166 @@
|
||||
"""
|
||||
Comprehensive tests of diffusion features that are available in online serving mode
|
||||
and are supported by the FluxKontext model.
|
||||
"""
|
||||
|
||||
import pytest
|
||||
|
||||
from tests.conftest import (
|
||||
OmniServer,
|
||||
OmniServerParams,
|
||||
OpenAIClientHandler,
|
||||
dummy_messages_from_mix_data,
|
||||
generate_synthetic_image,
|
||||
)
|
||||
|
||||
EDIT_PROMPT = "Transform this modern, geometrist image into a Vincent van Gogh style impressionist painting."
|
||||
NEGATIVE_PROMPT = "blurry, low quality, modern, geometrist"
|
||||
MODEL = "black-forest-labs/FLUX.1-Kontext-dev"
|
||||
|
||||
|
||||
def _get_diffusion_feature_cases(model: str):
|
||||
return [
|
||||
pytest.param(
|
||||
OmniServerParams(
|
||||
model=model,
|
||||
server_args=[
|
||||
"--tensor-parallel-size",
|
||||
"2",
|
||||
"--enable-cpu-offload",
|
||||
],
|
||||
),
|
||||
id="parallel_001",
|
||||
),
|
||||
]
|
||||
|
||||
|
||||
@pytest.mark.advanced_model
|
||||
@pytest.mark.diffusion
|
||||
@pytest.mark.parametrize(
|
||||
"omni_server",
|
||||
_get_diffusion_feature_cases(MODEL),
|
||||
indirect=True,
|
||||
)
|
||||
def test_flux_kontext_text_to_image(omni_server: OmniServer, openai_client: OpenAIClientHandler):
|
||||
"""Test text-to-image generation with FluxKontext in regular end-user scenarios."""
|
||||
messages = dummy_messages_from_mix_data(content_text="A photo of a cat sitting on a laptop keyboard")
|
||||
|
||||
request_config = {
|
||||
"model": omni_server.model,
|
||||
"messages": messages,
|
||||
"extra_body": {
|
||||
"height": 512,
|
||||
"width": 512,
|
||||
"num_inference_steps": 2,
|
||||
"seed": 42,
|
||||
},
|
||||
}
|
||||
|
||||
openai_client.send_diffusion_request(request_config)
|
||||
|
||||
|
||||
@pytest.mark.advanced_model
|
||||
@pytest.mark.diffusion
|
||||
@pytest.mark.parametrize(
|
||||
"omni_server",
|
||||
_get_diffusion_feature_cases(MODEL),
|
||||
indirect=True,
|
||||
)
|
||||
def test_flux_kontext_image_edit(omni_server: OmniServer, openai_client: OpenAIClientHandler):
|
||||
"""Test image editing with FluxKontext in regular end-user scenarios."""
|
||||
image_data_url = f"data:image/jpeg;base64,{generate_synthetic_image(512, 512)['base64']}"
|
||||
|
||||
messages = dummy_messages_from_mix_data(image_data_url=image_data_url, content_text=EDIT_PROMPT)
|
||||
|
||||
request_config = {
|
||||
"model": omni_server.model,
|
||||
"messages": messages,
|
||||
"extra_body": {
|
||||
"height": 512,
|
||||
"width": 512,
|
||||
"num_inference_steps": 2,
|
||||
"negative_prompt": NEGATIVE_PROMPT,
|
||||
"true_cfg_scale": 4.0,
|
||||
"seed": 42,
|
||||
},
|
||||
}
|
||||
|
||||
openai_client.send_diffusion_request(request_config)
|
||||
|
||||
|
||||
@pytest.mark.advanced_model
|
||||
@pytest.mark.diffusion
|
||||
@pytest.mark.parametrize(
|
||||
"omni_server",
|
||||
_get_diffusion_feature_cases(MODEL),
|
||||
indirect=True,
|
||||
)
|
||||
def test_flux_kontext_image_edit_no_negative(omni_server: OmniServer, openai_client: OpenAIClientHandler):
|
||||
"""Test image editing with FluxKontext without negative prompt."""
|
||||
image_data_url = f"data:image/jpeg;base64,{generate_synthetic_image(512, 512)['base64']}"
|
||||
|
||||
messages = dummy_messages_from_mix_data(image_data_url=image_data_url, content_text=EDIT_PROMPT)
|
||||
|
||||
request_config = {
|
||||
"model": omni_server.model,
|
||||
"messages": messages,
|
||||
"extra_body": {
|
||||
"height": 512,
|
||||
"width": 512,
|
||||
"num_inference_steps": 2,
|
||||
"seed": 42,
|
||||
},
|
||||
}
|
||||
|
||||
openai_client.send_diffusion_request(request_config)
|
||||
|
||||
|
||||
@pytest.mark.advanced_model
|
||||
@pytest.mark.diffusion
|
||||
@pytest.mark.parametrize(
|
||||
"omni_server",
|
||||
_get_diffusion_feature_cases(MODEL),
|
||||
indirect=True,
|
||||
)
|
||||
def test_flux_kontext_high_resolution(omni_server: OmniServer, openai_client: OpenAIClientHandler):
|
||||
"""Test high-resolution generation with FluxKontext."""
|
||||
messages = dummy_messages_from_mix_data(content_text="A beautiful landscape with mountains and a lake")
|
||||
|
||||
request_config = {
|
||||
"model": omni_server.model,
|
||||
"messages": messages,
|
||||
"extra_body": {
|
||||
"height": 768,
|
||||
"width": 1024,
|
||||
"num_inference_steps": 2,
|
||||
"seed": 42,
|
||||
},
|
||||
}
|
||||
|
||||
openai_client.send_diffusion_request(request_config)
|
||||
|
||||
|
||||
@pytest.mark.advanced_model
|
||||
@pytest.mark.diffusion
|
||||
@pytest.mark.parametrize(
|
||||
"omni_server",
|
||||
_get_diffusion_feature_cases(MODEL),
|
||||
indirect=True,
|
||||
)
|
||||
def test_flux_kontext_multiple_outputs(omni_server: OmniServer, openai_client: OpenAIClientHandler):
|
||||
"""Test generating multiple outputs with FluxKontext."""
|
||||
messages = dummy_messages_from_mix_data(content_text="A photo of a cat sitting on a laptop keyboard")
|
||||
|
||||
request_config = {
|
||||
"model": omni_server.model,
|
||||
"messages": messages,
|
||||
"extra_body": {
|
||||
"height": 512,
|
||||
"width": 512,
|
||||
"num_inference_steps": 2,
|
||||
"num_outputs_per_prompt": 2,
|
||||
"seed": 42,
|
||||
},
|
||||
}
|
||||
|
||||
openai_client.send_diffusion_request(request_config)
|
||||
Reference in new issue
Block a user