theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-23 17:38:10 +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

34
examples/wanvideo/model_inference/Wan2.2-VACE-Fun-A14B.py
View File

@@ -1,21 +1,35 @@
# Without VRAM Management, 80G VRAM is not enough to run this example.
# We recommend to use `examples/wanvideo/model_inference_low_vram/Wan2.2-VACE-Fun-A14B.py`.
# CPU Offload is enabled in this example.
import torch
from PIL import Image
from diffsynth import save_video, VideoData
from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
from diffsynth.utils.data import save_video, VideoData
from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
from modelscope import dataset_snapshot_download
vram_config = {
"offload_dtype": torch.bfloat16,
"offload_device": "cpu",
"onload_dtype": torch.bfloat16,
"onload_device": "cpu",
"preparing_dtype": torch.bfloat16,
"preparing_device": "cuda",
"computation_dtype": torch.bfloat16,
"computation_device": "cuda",
}
pipe = WanVideoPipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="high_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"),
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="low_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"),
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="high_noise_model/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="low_noise_model/diffusion_pytorch_model*.safetensors", **vram_config),
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
ModelConfig(model_id="PAI/Wan2.2-VACE-Fun-A14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
],
tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="google/umt5-xxl/"),
vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 2,
)
pipe.enable_vram_management()
dataset_snapshot_download(
@@ -32,7 +46,7 @@ video = pipe(
vace_video=control_video,
seed=1, tiled=True
)
save_video(video, "video1_14b.mp4", fps=15, quality=5)
save_video(video, "video_1_Wan2.2-VACE-Fun-A14B.mp4", fps=15, quality=5)
# Reference image -> Video
video = pipe(
@@ -41,7 +55,7 @@ video = pipe(
vace_reference_image=Image.open("data/examples/wan/cat_fightning.jpg").resize((832, 480)),
seed=1, tiled=True
)
save_video(video, "video2_14b.mp4", fps=15, quality=5)
save_video(video, "video_2_Wan2.2-VACE-Fun-A14B.mp4", fps=15, quality=5)
# Depth video + Reference image -> Video
video = pipe(
@@ -51,4 +65,4 @@ video = pipe(
vace_reference_image=Image.open("data/examples/wan/cat_fightning.jpg").resize((832, 480)),
seed=1, tiled=True
)
save_video(video, "video3_14b.mp4", fps=15, quality=5)
save_video(video, "video_3_Wan2.2-VACE-Fun-A14B.mp4", fps=15, quality=5)