theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-22 16:50:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

DiffSynth-Studio 2.0 major update

Browse Source
This commit is contained in:
root
2025-12-04 16:33:07 +08:00
parent afd101f345
commit 72af7122b3
758 changed files with 26462 additions and 2221398 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
examples/qwen_image/model_inference_low_vram/Qwen-Image-Blockwise-ControlNet-Canny.py
View File

@@ -4,18 +4,28 @@ import torch
from modelscope import dataset_snapshot_download
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Blockwise-ControlNet-Canny", origin_file_pattern="model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Blockwise-ControlNet-Canny", origin_file_pattern="model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
dataset_snapshot_download(
dataset_id="DiffSynth-Studio/example_image_dataset",

20
examples/qwen_image/model_inference_low_vram/Qwen-Image-Blockwise-ControlNet-Depth.py
View File

@@ -4,18 +4,28 @@ import torch
from modelscope import dataset_snapshot_download
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Blockwise-ControlNet-Depth", origin_file_pattern="model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Blockwise-ControlNet-Depth", origin_file_pattern="model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
dataset_snapshot_download(
dataset_id="DiffSynth-Studio/example_image_dataset",

20
examples/qwen_image/model_inference_low_vram/Qwen-Image-Blockwise-ControlNet-Inpaint.py
View File

@@ -4,18 +4,28 @@ from modelscope import dataset_snapshot_download
from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig, ControlNetInput
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Blockwise-ControlNet-Inpaint", origin_file_pattern="model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Blockwise-ControlNet-Inpaint", origin_file_pattern="model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
dataset_snapshot_download(
dataset_id="DiffSynth-Studio/example_image_dataset",

36
examples/qwen_image/model_inference_low_vram/Qwen-Image-Distill-DMD2.py Normal file
View File

@@ -0,0 +1,36 @@
from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
from diffsynth.core import load_state_dict
from modelscope import snapshot_download
import torch, math
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn, # bfloat16 is recommended.
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn, # bfloat16 is recommended.
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
snapshot_download("MusePublic/Qwen-Image-Distill", allow_file_pattern="qwen_image_distill_3step.safetensors", cache_dir="models")
lora_state_dict = load_state_dict("models/MusePublic/Qwen-Image-Distill/qwen_image_distill_3step.safetensors", device="cuda", torch_dtype=torch.bfloat16)
lora_state_dict = {i.replace("base_model.model.", "").replace(".weight", ".default.weight"): j for i, j in lora_state_dict.items()}
pipe.load_lora(pipe.dit, state_dict=lora_state_dict, hotload=True)
prompt = "精致肖像，水下少女，蓝裙飘逸，发丝轻扬，光影透澈，气泡环绕，面容恬静，细节精致，梦幻唯美。"
image = pipe(prompt, seed=0, num_inference_steps=3, cfg_scale=1, exponential_shift_mu=math.log(2.5))
image.save("image.jpg")

18
examples/qwen_image/model_inference_low_vram/Qwen-Image-Distill-Full.py
View File

@@ -2,17 +2,27 @@ from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
import torch
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Distill-Full", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="DiffSynth-Studio/Qwen-Image-Distill-Full", origin_file_pattern="diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
prompt = "精致肖像，水下少女，蓝裙飘逸，发丝轻扬，光影透澈，气泡环绕，面容恬静，细节精致，梦幻唯美。"
image = pipe(prompt, seed=0, num_inference_steps=15, cfg_scale=1)
image.save("image.jpg")

21
examples/qwen_image/model_inference_low_vram/Qwen-Image-Distill-LoRA.py
View File

@@ -2,20 +2,29 @@ from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
from modelscope import snapshot_download
import torch
# Please do not use float8 on this model
snapshot_download("DiffSynth-Studio/Qwen-Image-Distill-LoRA", local_dir="models/DiffSynth-Studio/Qwen-Image-Distill-LoRA")
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu"),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu"),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu"),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-Distill-LoRA/model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-Distill-LoRA/model.safetensors", hotload=True)
prompt = "精致肖像，水下少女，蓝裙飘逸，发丝轻扬，光影透澈，气泡环绕，面容恬静，细节精致，梦幻唯美。"
image = pipe(prompt, seed=0, num_inference_steps=15, cfg_scale=1)

25
examples/qwen_image/model_inference_low_vram/Qwen-Image-Edit-2509.py
View File

@@ -2,17 +2,28 @@ from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
from PIL import Image
import torch
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image-Edit-2509", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image-Edit-2509", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
processor_config=ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="processor/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
image_1 = pipe(prompt="一位少女", seed=0, num_inference_steps=40, height=1328, width=1024)
image_1.save("image1.jpg")
@@ -24,3 +35,9 @@ prompt = "生成这两个人的合影"
edit_image = [Image.open("image1.jpg"), Image.open("image2.jpg")]
image_3 = pipe(prompt, edit_image=edit_image, seed=1, num_inference_steps=40, height=1328, width=1024, edit_image_auto_resize=True)
image_3.save("image3.jpg")
# Qwen-Image-Edit-2509 is a multi-image editing model.
# Please use a list to input `edit_image`, even if the input contains only one image.
# edit_image = [Image.open("image.jpg")]
# Please do not input the image directly.
# edit_image = Image.open("image.jpg")

22
examples/qwen_image/model_inference_low_vram/Qwen-Image-Edit-Lowres-Fix.py
View File

@@ -2,20 +2,30 @@ from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
import torch
from modelscope import snapshot_download
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
processor_config=ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="processor/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
snapshot_download("DiffSynth-Studio/Qwen-Image-Edit-Lowres-Fix", local_dir="models/DiffSynth-Studio/Qwen-Image-Edit-Lowres-Fix", allow_file_pattern="model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-Edit-Lowres-Fix/model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-Edit-Lowres-Fix/model.safetensors", hotload=True)
prompt = "精致肖像，水下少女，蓝裙飘逸，发丝轻扬，光影透澈，气泡环绕，面容恬静，细节精致，梦幻唯美。"
image = pipe(prompt=prompt, seed=0, num_inference_steps=40, height=1024, width=768)

31
examples/qwen_image/model_inference_low_vram/Qwen-Image-Edit.py
View File

@@ -1,22 +1,37 @@
from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
import torch
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
processor_config=ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="processor/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
prompt = "精致肖像，水下少女，蓝裙飘逸，发丝轻扬，光影透澈，气泡环绕，面容恬静，细节精致，梦幻唯美。"
image = pipe(prompt=prompt, seed=0, num_inference_steps=40, height=1024, width=1024)
image.save("image1.jpg")
input_image = pipe(prompt=prompt, seed=0, num_inference_steps=40, height=1328, width=1024)
input_image.save("image1.jpg")
prompt = "将裙子改为粉色"
image = pipe(prompt, edit_image=image, seed=1, num_inference_steps=40, height=1024, width=1024)
# edit_image_auto_resize=True: auto resize input image to match the area of 1024*1024 with the original aspect ratio
image = pipe(prompt, edit_image=input_image, seed=1, num_inference_steps=40, height=1328, width=1024, edit_image_auto_resize=True)
image.save(f"image2.jpg")
# edit_image_auto_resize=False: do not resize input image
image = pipe(prompt, edit_image=input_image, seed=1, num_inference_steps=40, height=1328, width=1024, edit_image_auto_resize=False)
image.save(f"image3.jpg")

20
examples/qwen_image/model_inference_low_vram/Qwen-Image-EliGen-Poster.py
View File

@@ -92,23 +92,33 @@ def example(pipe, seeds, example_id, global_prompt, entity_prompts, height=784,
visualize_masks(image, masks, entity_prompts, f"eligen_poster_example_{example_id}_mask_{seed}.png")
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
snapshot_download(
"DiffSynth-Studio/Qwen-Image-EliGen-Poster",
local_dir="models/DiffSynth-Studio/Qwen-Image-EliGen-Poster",
allow_file_pattern="model.safetensors",
)
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-EliGen-Poster/model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-EliGen-Poster/model.safetensors", hotload=True)
global_prompt = "一张以柔粉紫为背景的海报，左侧有大号粉紫色文字“Qwen-Image EliGen-Poster”，粉紫色椭圆框内白色小字：“图像精确分区控制模型”。右侧有一只小兔子在拆礼物，旁边站着一只头顶迷你烟花发射器的小龙（卡通Q版）。背景有一些白云点缀。整体风格卡通可爱，传达节日惊喜的主题。"
entity_prompts = ["粉紫色文字“Qwen-Image EliGen-Poster”", "粉紫色椭圆框内白色小字：“图像精确分区控制模型”", "一只小兔子在拆礼物，小兔子旁边站着一只头顶迷你烟花发射器的小龙（卡通Q版）"]
seed = [42]

21
examples/qwen_image/model_inference_low_vram/Qwen-Image-EliGen-V2.py
View File

@@ -4,7 +4,6 @@ from PIL import Image, ImageDraw, ImageFont
from modelscope import dataset_snapshot_download, snapshot_download
from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
def visualize_masks(image, masks, mask_prompts, output_path, font_size=35, use_random_colors=False):
# Create a blank image for overlays
overlay = Image.new('RGBA', image.size, (0, 0, 0, 0))
@@ -83,19 +82,29 @@ def example(pipe, seeds, example_id, global_prompt, entity_prompts):
visualize_masks(image, masks, entity_prompts, f"eligen_example_{example_id}_mask_{seed}.png")
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
snapshot_download("DiffSynth-Studio/Qwen-Image-EliGen-V2", local_dir="models/DiffSynth-Studio/Qwen-Image-EliGen-V2", allow_file_pattern="model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-EliGen-V2/model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-EliGen-V2/model.safetensors", hotload=True)
seeds = [0]

41
examples/qwen_image/model_inference_low_vram/Qwen-Image-EliGen.py
View File

@@ -83,19 +83,29 @@ def example(pipe, seeds, example_id, global_prompt, entity_prompts):
visualize_masks(image, masks, entity_prompts, f"eligen_example_{example_id}_mask_{seed}.png")
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
snapshot_download("DiffSynth-Studio/Qwen-Image-EliGen", local_dir="models/DiffSynth-Studio/Qwen-Image-EliGen", allow_file_pattern="model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-EliGen/model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-EliGen/model.safetensors", hotload=True)
# example 1
global_prompt = "A breathtaking beauty of Raja Ampat by the late-night moonlight , one beautiful woman from behind wearing a pale blue long dress with soft glow, sitting at the top of a cliff looking towards the beach,pastell light colors, a group of small distant birds flying in far sky, a boat sailing on the sea, best quality, realistic, whimsical, fantastic, splash art, intricate detailed, hyperdetailed, maximalist style, photorealistic, concept art, sharp focus, harmony, serenity, tranquility, soft pastell colors,ambient occlusion, cozy ambient lighting, masterpiece, liiv1, linquivera, metix, mentixis, masterpiece, award winning, view from above\n"
@@ -106,24 +116,3 @@ example(pipe, [0], 1, global_prompt, entity_prompts)
global_prompt = "samurai girl wearing a kimono, she's holding a sword glowing with red flame, her long hair is flowing in the wind, she is looking at a small bird perched on the back of her hand. ultra realist style. maximum image detail. maximum realistic render."
entity_prompts = ["flowing hair", "sword glowing with red flame", "A cute bird", "yellow belt"]
example(pipe, [0], 2, global_prompt, entity_prompts)
# example 3
global_prompt = "Image of a neverending staircase up to a mysterious palace in the sky, The ancient palace stood majestically atop a mist-shrouded mountain, sunrise, two traditional monk walk in the stair looking at the sunrise, fog,see-through, best quality, whimsical, fantastic, splash art, intricate detailed, hyperdetailed, photorealistic, concept art, harmony, serenity, tranquility, ambient occlusion, halation, cozy ambient lighting, dynamic lighting,masterpiece, liiv1, linquivera, metix, mentixis, masterpiece, award winning,"
entity_prompts = ["ancient palace", "stone staircase with railings", "a traditional monk", "a traditional monk"]
example(pipe, [27], 3, global_prompt, entity_prompts)
# example 4
global_prompt = "A beautiful girl wearing shirt and shorts in the street, holding a sign 'Entity Control'"
entity_prompts = ["A beautiful girl", "sign 'Entity Control'", "shorts", "shirt"]
example(pipe, [21], 4, global_prompt, entity_prompts)
# example 5
global_prompt = "A captivating, dramatic scene in a painting that exudes mystery and foreboding. A white sky, swirling blue clouds, and a crescent yellow moon illuminate a solitary woman standing near the water's edge. Her long dress flows in the wind, silhouetted against the eerie glow. The water mirrors the fiery sky and moonlight, amplifying the uneasy atmosphere."
entity_prompts = ["crescent yellow moon", "a solitary woman", "water", "swirling blue clouds"]
example(pipe, [0], 5, global_prompt, entity_prompts)
# example 7, same prompt with different seeds
seeds = range(5, 9)
global_prompt = "A beautiful asia woman wearing white dress, holding a mirror, with a forest background."
entity_prompts = ["A beautiful woman", "mirror", "necklace", "glasses", "earring", "white dress", "jewelry headpiece"]
example(pipe, seeds, 7, global_prompt, entity_prompts)

22
examples/qwen_image/model_inference_low_vram/Qwen-Image-In-Context-Control-Union.py
View File

@@ -2,24 +2,34 @@ from PIL import Image
import torch
from modelscope import dataset_snapshot_download, snapshot_download
from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
from diffsynth.controlnets.processors import Annotator
from diffsynth.utils.controlnet import Annotator
allow_file_pattern = ["sk_model.pth", "sk_model2.pth", "dpt_hybrid-midas-501f0c75.pt", "ControlNetHED.pth", "body_pose_model.pth", "hand_pose_model.pth", "facenet.pth", "scannet.pt"]
snapshot_download("lllyasviel/Annotators", local_dir="models/Annotators", allow_file_pattern=allow_file_pattern)
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
snapshot_download("DiffSynth-Studio/Qwen-Image-In-Context-Control-Union", local_dir="models/DiffSynth-Studio/Qwen-Image-In-Context-Control-Union", allow_file_pattern="model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-In-Context-Control-Union/model.safetensors")
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/Qwen-Image-In-Context-Control-Union/model.safetensors", hotload=True)
dataset_snapshot_download(dataset_id="DiffSynth-Studio/examples_in_diffsynth", local_dir="./", allow_file_pattern=f"data/examples/qwen-image-context-control/image.jpg")
origin_image = Image.open("data/examples/qwen-image-context-control/image.jpg").resize((1024, 1024))

18
examples/qwen_image/model_inference_low_vram/Qwen-Image.py
View File

@@ -2,17 +2,27 @@ from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
import torch
vram_config = {
"offload_dtype": "disk",
"offload_device": "disk",
"onload_dtype": torch.float8_e4m3fn,
"onload_device": "cpu",
"preparing_dtype": torch.float8_e4m3fn,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = QwenImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="tokenizer/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
pipe.enable_vram_management()
prompt = "精致肖像，水下少女，蓝裙飘逸，发丝轻扬，光影透澈，气泡环绕，面容恬静，细节精致，梦幻唯美。"
image = pipe(prompt, seed=0, num_inference_steps=40)
image.save("image.jpg")