From 46bd05b54d8aef87f81d70c4714c8b68f0d4c388 Mon Sep 17 00:00:00 2001 From: "lzw478614@alibaba-inc.com" Date: Thu, 21 Aug 2025 13:41:07 +0800 Subject: [PATCH] add examples --- .../Wan2.2-Fun-A14B-Control-Camera.py | 43 +++++++++++++++++++ .../Wan2.2-Fun-A14B-Control.py | 35 +++++++++++++++ .../model_inference/Wan2.2-Fun-A14B-Inp.py | 35 +++++++++++++++ 3 files changed, 113 insertions(+) create mode 100644 examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py create mode 100644 examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py create mode 100644 examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Inp.py diff --git a/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py b/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py new file mode 100644 index 0000000..27cda27 --- /dev/null +++ b/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py @@ -0,0 +1,43 @@ +import torch +from diffsynth import save_video,VideoData +from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig +from PIL import Image +from modelscope import dataset_snapshot_download + +pipe = WanVideoPipeline.from_pretrained( + torch_dtype=torch.bfloat16, + device="cuda", + model_configs=[ + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control-Camera", origin_file_pattern="high_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control-Camera", origin_file_pattern="low_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control-Camera", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control-Camera", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"), + ], +) +pipe.enable_vram_management() + + +dataset_snapshot_download( + dataset_id="DiffSynth-Studio/examples_in_diffsynth", + local_dir="./", + allow_file_pattern=f"data/examples/wan/input_image.jpg" +) +input_image = Image.open("data/examples/wan/input_image.jpg") + +video = pipe( + prompt="一艘小船正勇敢地乘风破浪前行。蔚蓝的大海波涛汹涌，白色的浪花拍打着船身，但小船毫不畏惧，坚定地驶向远方。阳光洒在水面上，闪烁着金色的光芒，为这壮丽的场景增添了一抹温暖。镜头拉近，可以看到船上的旗帜迎风飘扬，象征着不屈的精神与冒险的勇气。这段画面充满力量，激励人心，展现了面对挑战时的无畏与执着。", + negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走", + seed=0, tiled=True, + input_image=input_image, + camera_control_direction="Left", camera_control_speed=0.01, +) +save_video(video, "video_left.mp4", fps=15, quality=5) + +video = pipe( + prompt="一艘小船正勇敢地乘风破浪前行。蔚蓝的大海波涛汹涌，白色的浪花拍打着船身，但小船毫不畏惧，坚定地驶向远方。阳光洒在水面上，闪烁着金色的光芒，为这壮丽的场景增添了一抹温暖。镜头拉近，可以看到船上的旗帜迎风飘扬，象征着不屈的精神与冒险的勇气。这段画面充满力量，激励人心，展现了面对挑战时的无畏与执着。", + negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走", + seed=0, tiled=True, + input_image=input_image, + camera_control_direction="Up", camera_control_speed=0.01, +) +save_video(video, "video_up.mp4", fps=15, quality=5) \ No newline at end of file diff --git a/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py b/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py new file mode 100644 index 0000000..2941422 --- /dev/null +++ b/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py @@ -0,0 +1,35 @@ +import torch +from diffsynth import save_video,VideoData +from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig +from PIL import Image +from modelscope import dataset_snapshot_download + +pipe = WanVideoPipeline.from_pretrained( + torch_dtype=torch.bfloat16, + device="cuda", + model_configs=[ + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control", origin_file_pattern="high_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control", origin_file_pattern="low_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-Control", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"), + ], +) +pipe.enable_vram_management() + +dataset_snapshot_download( + dataset_id="DiffSynth-Studio/examples_in_diffsynth", + local_dir="./", + allow_file_pattern=["data/examples/wan/control_video.mp4", "data/examples/wan/reference_image_girl.png"] +) + +# Control video +control_video = VideoData("data/examples/wan/control_video.mp4", height=832, width=576) +reference_image = Image.open("data/examples/wan/reference_image_girl.png").resize((576, 832)) +video = pipe( + prompt="扁平风格动漫，一位长发少女优雅起舞。她五官精致，大眼睛明亮有神，黑色长发柔顺光泽。身穿淡蓝色T恤和深蓝色牛仔短裤。背景是粉色。", + negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走", + control_video=control_video, reference_image=reference_image, + height=832, width=576, num_frames=49, + seed=1, tiled=True +) +save_video(video, "video.mp4", fps=15, quality=5) \ No newline at end of file diff --git a/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Inp.py b/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Inp.py new file mode 100644 index 0000000..87c8488 --- /dev/null +++ b/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Inp.py @@ -0,0 +1,35 @@ +import torch +from diffsynth import save_video +from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig +from PIL import Image +from modelscope import dataset_snapshot_download + +pipe = WanVideoPipeline.from_pretrained( + torch_dtype=torch.bfloat16, + device="cuda", + model_configs=[ + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-InP", origin_file_pattern="high_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-InP", origin_file_pattern="low_noise_model/diffusion_pytorch_model*.safetensors", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-InP", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"), + ModelConfig(model_id="PAI/Wan2.2-Fun-A14B-InP", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"), + ], +) +pipe.enable_vram_management() + +dataset_snapshot_download( + dataset_id="DiffSynth-Studio/examples_in_diffsynth", + local_dir="./", + allow_file_pattern=f"data/examples/wan/input_image.jpg" +) +image = Image.open("data/examples/wan/input_image.jpg") + +# First and last frame to video +video = pipe( + prompt="一艘小船正勇敢地乘风破浪前行。蔚蓝的大海波涛汹涌，白色的浪花拍打着船身，但小船毫不畏惧，坚定地驶向远方。阳光洒在水面上，闪烁着金色的光芒，为这壮丽的场景增添了一抹温暖。镜头拉近，可以看到船上的旗帜迎风飘扬，象征着不屈的精神与冒险的勇气。这段画面充满力量，激励人心，展现了面对挑战时的无畏与执着。", + negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走", + input_image=image, + seed=0, tiled=True, + # You can input `end_image=xxx` to control the last frame of the video. + # The model will automatically generate the dynamic content between `input_image` and `end_image`. +) +save_video(video, "video.mp4", fps=15, quality=5) \ No newline at end of file