Diffusers documentation

Wan

Diffusers

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v0.32.2).

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

Wan

Wan 2.1 by the Alibaba Wan Team.

Generating Videos with Wan 2.1

We will first need to install some addtional dependencies.

pip install -u ftfy imageio-ffmpeg imageio

Text to Video Generation

The following example requires 11GB VRAM to run and uses the smaller Wan-AI/Wan2.1-T2V-1.3B-Diffusers model. You can switch it out for the larger Wan2.1-I2V-14B-720P-Diffusers or Wan-AI/Wan2.1-I2V-14B-480P-Diffusers if you have at least 35GB VRAM available.

from diffusers import WanPipeline
from diffusers.utils import export_to_video

# Available models: Wan-AI/Wan2.1-I2V-14B-720P-Diffusers or Wan-AI/Wan2.1-I2V-14B-480P-Diffusers
model_id = "Wan-AI/Wan2.1-T2V-1.3B-Diffusers"

pipe = WanPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16)
pipe.enable_model_cpu_offload()

prompt = "A cat and a dog baking a cake together in a kitchen. The cat is carefully measuring flour, while the dog is stirring the batter with a wooden spoon. The kitchen is cozy, with sunlight streaming through the window."
negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"
num_frames = 33

frames = pipe(prompt=prompt, negative_prompt=negative_prompt, num_frames=num_frames).frames[0]
export_to_video(frames, "wan-t2v.mp4", fps=16)

You can improve the quality of the generated video by running the decoding step in full precision.

from diffusers import WanPipeline, AutoencoderKLWan
from diffusers.utils import export_to_video

model_id = "Wan-AI/Wan2.1-T2V-1.3B-Diffusers"

vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)
pipe = WanPipeline.from_pretrained(model_id, vae=vae, torch_dtype=torch.bfloat16)

# replace this with pipe.to("cuda") if you have sufficient VRAM
pipe.enable_model_cpu_offload()

prompt = "A cat and a dog baking a cake together in a kitchen. The cat is carefully measuring flour, while the dog is stirring the batter with a wooden spoon. The kitchen is cozy, with sunlight streaming through the window."
negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"
num_frames = 33

frames = pipe(prompt=prompt, num_frames=num_frames).frames[0]
export_to_video(frames, "wan-t2v.mp4", fps=16)

Image to Video Generation

The Image to Video pipeline requires loading the AutoencoderKLWan and the CLIPVisionModel components in full precision. The following example will need at least 35GB of VRAM to run.

import torch
import numpy as np
from diffusers import AutoencoderKLWan, WanImageToVideoPipeline
from diffusers.utils import export_to_video, load_image
from transformers import CLIPVisionModel

# Available models: Wan-AI/Wan2.1-I2V-14B-480P-Diffusers, Wan-AI/Wan2.1-I2V-14B-720P-Diffusers
model_id = "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers"
image_encoder = CLIPVisionModel.from_pretrained(
    model_id, subfolder="image_encoder", torch_dtype=torch.float32
)
vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)
pipe = WanImageToVideoPipeline.from_pretrained(
    model_id, vae=vae, image_encoder=image_encoder, torch_dtype=torch.bfloat16
)

# replace this with pipe.to("cuda") if you have sufficient VRAM
pipe.enable_model_cpu_offload()

image = load_image(
    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astronaut.jpg"
)

max_area = 480 * 832
aspect_ratio = image.height / image.width
mod_value = pipe.vae_scale_factor_spatial * pipe.transformer.config.patch_size[1]
height = round(np.sqrt(max_area * aspect_ratio)) // mod_value * mod_value
width = round(np.sqrt(max_area / aspect_ratio)) // mod_value * mod_value
image = image.resize((width, height))

prompt = (
    "An astronaut hatching from an egg, on the surface of the moon, the darkness and depth of space realised in "
    "the background. High quality, ultrarealistic detail and breath-taking movie-like camera shot."
)
negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"

num_frames = 33

output = pipe(
    image=image,
    prompt=prompt,
    negative_prompt=negative_prompt,
    height=height,
    width=width,
    num_frames=num_frames,
    guidance_scale=5.0,
).frames[0]
export_to_video(output, "wan-i2v.mp4", fps=16)

Video to Video Generation

import torch
from diffusers.utils import load_video, export_to_video
from diffusers import AutoencoderKLWan, WanVideoToVideoPipeline, UniPCMultistepScheduler

# Available models: Wan-AI/Wan2.1-T2V-14B-Diffusers, Wan-AI/Wan2.1-T2V-1.3B-Diffusers
model_id = "Wan-AI/Wan2.1-T2V-1.3B-Diffusers"
vae = AutoencoderKLWan.from_pretrained(
    model_id, subfolder="vae", torch_dtype=torch.float32
)
pipe = WanVideoToVideoPipeline.from_pretrained(
    model_id, vae=vae, torch_dtype=torch.bfloat16
)
flow_shift = 3.0  # 5.0 for 720P, 3.0 for 480P
pipe.scheduler = UniPCMultistepScheduler.from_config(
    pipe.scheduler.config, flow_shift=flow_shift
)
# change to pipe.to("cuda") if you have sufficient VRAM
pipe.enable_model_cpu_offload()

prompt = "A robot standing on a mountain top. The sun is setting in the background"
negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"
video = load_video(
    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/hiker.mp4"
)
output = pipe(
    video=video,
    prompt=prompt,
    negative_prompt=negative_prompt,
    height=480,
    width=512,
    guidance_scale=7.0,
    strength=0.7,
).frames[0]

export_to_video(output, "wan-v2v.mp4", fps=16)

Memory Optimizations for Wan 2.1

Base inference with the large 14B Wan 2.1 models can take up to 35GB of VRAM when generating videos at 720p resolution. We’ll outline a few memory optimizations we can apply to reduce the VRAM required to run the model.

We’ll use Wan-AI/Wan2.1-I2V-14B-720P-Diffusers model in these examples to demonstrate the memory savings, but the techniques are applicable to all model checkpoints.

Group Offloading the Transformer and UMT5 Text Encoder

Find more information about group offloading here

Block Level Group Offloading

We can reduce our VRAM requirements by applying group offloading to the larger model components of the pipeline; the WanTransformer3DModel and UMT5EncoderModel. Group offloading will break up the individual modules of a model and offload/onload them onto your GPU as needed during inference. In this example, we’ll apply block_level offloading, which will group the modules in a model into blocks of size num_blocks_per_group and offload/onload them to GPU. Moving to between CPU and GPU does add latency to the inference process. You can trade off between latency and memory savings by increasing or decreasing the num_blocks_per_group.

The following example will now only require 14GB of VRAM to run, but will take approximately 30 minutes to generate a video.

import torch
import numpy as np
from diffusers import AutoencoderKLWan, WanTransformer3DModel, WanImageToVideoPipeline
from diffusers.hooks.group_offloading import apply_group_offloading
from diffusers.utils import export_to_video, load_image
from transformers import UMT5EncoderModel, CLIPVisionModel

# Available models: Wan-AI/Wan2.1-I2V-14B-480P-Diffusers, Wan-AI/Wan2.1-I2V-14B-720P-Diffusers
model_id = "Wan-AI/Wan2.1-I2V-14B-720P-Diffusers"
image_encoder = CLIPVisionModel.from_pretrained(
    model_id, subfolder="image_encoder", torch_dtype=torch.float32
)

text_encoder = UMT5EncoderModel.from_pretrained(model_id, subfolder="text_encoder", torch_dtype=torch.bfloat16)
vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)
transformer = WanTransformer3DModel.from_pretrained(model_id, subfolder="transformer", torch_dtype=torch.bfloat16)

onload_device = torch.device("cuda")
offload_device = torch.device("cpu")

apply_group_offloading(text_encoder,
    onload_device=onload_device,
    offload_device=offload_device,
    offload_type="block_level",
    num_blocks_per_group=4
)

transformer.enable_group_offload(
    onload_device=onload_device,
    offload_device=offload_device,
    offload_type="block_level",
    num_blocks_per_group=4,
)
pipe = WanImageToVideoPipeline.from_pretrained(
    model_id,
    vae=vae,
    transformer=transformer,
    text_encoder=text_encoder,
    image_encoder=image_encoder,
    torch_dtype=torch.bfloat16
)
# Since we've offloaded the larger models alrady, we can move the rest of the model components to GPU
pipe.to("cuda")

image = load_image(
    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astronaut.jpg"
)

max_area = 720 * 832
aspect_ratio = image.height / image.width
mod_value = pipe.vae_scale_factor_spatial * pipe.transformer.config.patch_size[1]
height = round(np.sqrt(max_area * aspect_ratio)) // mod_value * mod_value
width = round(np.sqrt(max_area / aspect_ratio)) // mod_value * mod_value
image = image.resize((width, height))

prompt = (
    "An astronaut hatching from an egg, on the surface of the moon, the darkness and depth of space realised in "
    "the background. High quality, ultrarealistic detail and breath-taking movie-like camera shot."
)
negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"

num_frames = 33

output = pipe(
    image=image,
    prompt=prompt,
    negative_prompt=negative_prompt,
    height=height,
    width=width,
    num_frames=num_frames,
    guidance_scale=5.0,
).frames[0]

export_to_video(output, "wan-i2v.mp4", fps=16)

Block Level Group Offloading with CUDA Streams

We can speed up group offloading inference, by enabling the use of CUDA streams. However, using CUDA streams requires moving the model parameters into pinned memory. This allocation is handled by Pytorch under the hood, and can result in a significant spike in CPU RAM usage. Please consider this option if your CPU RAM is atleast 2X the size of the model you are group offloading.

In the following example we will use CUDA streams when group offloading the WanTransformer3DModel. When testing on an A100, this example will require 14GB of VRAM, 52GB of CPU RAM, but will generate a video in approximately 9 minutes.

import torch
import numpy as np
from diffusers import AutoencoderKLWan, WanTransformer3DModel, WanImageToVideoPipeline
from diffusers.hooks.group_offloading import apply_group_offloading
from diffusers.utils import export_to_video, load_image
from transformers import UMT5EncoderModel, CLIPVisionModel

# Available models: Wan-AI/Wan2.1-I2V-14B-480P-Diffusers, Wan-AI/Wan2.1-I2V-14B-720P-Diffusers
model_id = "Wan-AI/Wan2.1-I2V-14B-720P-Diffusers"
image_encoder = CLIPVisionModel.from_pretrained(
    model_id, subfolder="image_encoder", torch_dtype=torch.float32
)

text_encoder = UMT5EncoderModel.from_pretrained(model_id, subfolder="text_encoder", torch_dtype=torch.bfloat16)
vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)
transformer = WanTransformer3DModel.from_pretrained(model_id, subfolder="transformer", torch_dtype=torch.bfloat16)

onload_device = torch.device("cuda")
offload_device = torch.device("cpu")

apply_group_offloading(text_encoder,
    onload_device=onload_device,
    offload_device=offload_device,
    offload_type="block_level",
    num_blocks_per_group=4
)

transformer.enable_group_offload(
    onload_device=onload_device,
    offload_device=offload_device,
    offload_type="leaf_level",
    use_stream=True
)
pipe = WanImageToVideoPipeline.from_pretrained(
    model_id,
    vae=vae,
    transformer=transformer,
    text_encoder=text_encoder,
    image_encoder=image_encoder,
    torch_dtype=torch.bfloat16
)
# Since we've offloaded the larger models alrady, we can move the rest of the model components to GPU
pipe.to("cuda")

image = load_image(
    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astronaut.jpg"
)

max_area = 720 * 832
aspect_ratio = image.height / image.width
mod_value = pipe.vae_scale_factor_spatial * pipe.transformer.config.patch_size[1]
height = round(np.sqrt(max_area * aspect_ratio)) // mod_value * mod_value
width = round(np.sqrt(max_area / aspect_ratio)) // mod_value * mod_value
image = image.resize((width, height))

prompt = (
    "An astronaut hatching from an egg, on the surface of the moon, the darkness and depth of space realised in "
    "the background. High quality, ultrarealistic detail and breath-taking movie-like camera shot."
)
negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"

num_frames = 33

output = pipe(
    image=image,
    prompt=prompt,
    negative_prompt=negative_prompt,
    height=height,
    width=width,
    num_frames=num_frames,
    guidance_scale=5.0,
).frames[0]

export_to_video(output, "wan-i2v.mp4", fps=16)

Applying Layerwise Casting to the Transformer

Find more information about layerwise casting here

In this example, we will model offloading with layerwise casting. Layerwise casting will downcast each layer’s weights to torch.float8_e4m3fn, temporarily upcast to torch.bfloat16 during the forward pass of the layer, then revert to torch.float8_e4m3fn afterward. This approach reduces memory requirements by approximately 50% while introducing a minor quality reduction in the generated video due to the precision trade-off.

This example will require 20GB of VRAM.

import torch
import numpy as np
from diffusers import AutoencoderKLWan, WanTransformer3DModel, WanImageToVideoPipeline
from diffusers.hooks.group_offloading import apply_group_offloading
from diffusers.utils import export_to_video, load_image
from transformers import UMT5EncoderModel, CLIPVisionModel

model_id = "Wan-AI/Wan2.1-I2V-14B-720P-Diffusers"
image_encoder = CLIPVisionModel.from_pretrained(
    model_id, subfolder="image_encoder", torch_dtype=torch.float32
)
text_encoder = UMT5EncoderModel.from_pretrained(model_id, subfolder="text_encoder", torch_dtype=torch.bfloat16)
vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)

transformer = WanTransformer3DModel.from_pretrained(model_id, subfolder="transformer", torch_dtype=torch.bfloat16)
transformer.enable_layerwise_casting(storage_dtype=torch.float8_e4m3fn, compute_dtype=torch.bfloat16)

pipe = WanImageToVideoPipeline.from_pretrained(
    model_id,
    vae=vae,
    transformer=transformer,
    text_encoder=text_encoder,
    image_encoder=image_encoder,
    torch_dtype=torch.bfloat16
)
pipe.enable_model_cpu_offload()
image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astronaut.jpg")

max_area = 720 * 832
aspect_ratio = image.height / image.width
mod_value = pipe.vae_scale_factor_spatial * pipe.transformer.config.patch_size[1]
height = round(np.sqrt(max_area * aspect_ratio)) // mod_value * mod_value
width = round(np.sqrt(max_area / aspect_ratio)) // mod_value * mod_value
image = image.resize((width, height))
prompt = (
    "An astronaut hatching from an egg, on the surface of the moon, the darkness and depth of space realised in "
    "the background. High quality, ultrarealistic detail and breath-taking movie-like camera shot."
)
negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"
num_frames = 33

output = pipe(
    image=image,
    prompt=prompt,
    negative_prompt=negative_prompt,
    height=height,
    width=width,
    num_frames=num_frames,
    num_inference_steps=50,
    guidance_scale=5.0,
).frames[0]
export_to_video(output, "wan-i2v.mp4", fps=16)

Using a Custom Scheduler

Wan can be used with many different schedulers, each with their own benefits regarding speed and generation quality. By default, Wan uses the UniPCMultistepScheduler(prediction_type="flow_prediction", use_flow_sigmas=True, flow_shift=3.0) scheduler. You can use a different scheduler as follows:

from diffusers import FlowMatchEulerDiscreteScheduler, UniPCMultistepScheduler, WanPipeline

scheduler_a = FlowMatchEulerDiscreteScheduler(shift=5.0)
scheduler_b = UniPCMultistepScheduler(prediction_type="flow_prediction", use_flow_sigmas=True, flow_shift=4.0)

pipe = WanPipeline.from_pretrained("Wan-AI/Wan2.1-T2V-1.3B-Diffusers", scheduler=<CUSTOM_SCHEDULER_HERE>)

# or,
pipe.scheduler = <CUSTOM_SCHEDULER_HERE>

Using Single File Loading with Wan 2.1

The WanTransformer3DModel and AutoencoderKLWan models support loading checkpoints in their original format via the from_single_file loading method.

import torch
from diffusers import WanPipeline, WanTransformer3DModel

ckpt_path = "https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/blob/main/split_files/diffusion_models/wan2.1_t2v_1.3B_bf16.safetensors"
transformer = WanTransformer3DModel.from_single_file(ckpt_path, torch_dtype=torch.bfloat16)

pipe = WanPipeline.from_pretrained("Wan-AI/Wan2.1-T2V-1.3B-Diffusers", transformer=transformer)

Recommendations for Inference

Keep AutencoderKLWan in torch.float32 for better decoding quality.
num_frames should satisfy the following constraint: (num_frames - 1) % 4 == 0
For smaller resolution videos, try lower values of shift (between 2.0 to 5.0) in the Scheduler. For larger resolution videos, try higher values (between 7.0 and 12.0). The default value is 3.0 for Wan.

WanPipeline

class diffusers.WanPipeline

< source >

( tokenizer: AutoTokenizer text_encoder: UMT5EncoderModel transformer: WanTransformer3DModel vae: AutoencoderKLWan scheduler: FlowMatchEulerDiscreteScheduler )

Parameters

tokenizer (T5Tokenizer) — Tokenizer from T5, specifically the google/umt5-xxl variant.
text_encoder (T5EncoderModel) — T5, specifically the google/umt5-xxl variant.
transformer (WanTransformer3DModel) — Conditional Transformer to denoise the input latents.
scheduler (UniPCMultistepScheduler) — A scheduler to be used in combination with transformer to denoise the encoded image latents.
vae (AutoencoderKLWan) — Variational Auto-Encoder (VAE) Model to encode and decode videos to and from latent representations.

Pipeline for text-to-video generation using Wan.

This model inherits from DiffusionPipeline. Check the superclass documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular device, etc.).

call

< source >

( prompt: typing.Union[str, typing.List[str]] = None negative_prompt: typing.Union[str, typing.List[str]] = None height: int = 480 width: int = 832 num_frames: int = 81 num_inference_steps: int = 50 guidance_scale: float = 5.0 num_videos_per_prompt: typing.Optional[int] = 1 generator: typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None latents: typing.Optional[torch.Tensor] = None prompt_embeds: typing.Optional[torch.Tensor] = None negative_prompt_embeds: typing.Optional[torch.Tensor] = None output_type: typing.Optional[str] = 'np' return_dict: bool = True attention_kwargs: typing.Optional[typing.Dict[str, typing.Any]] = None callback_on_step_end: typing.Union[typing.Callable[[int, int, typing.Dict], NoneType], diffusers.callbacks.PipelineCallback, diffusers.callbacks.MultiPipelineCallbacks, NoneType] = None callback_on_step_end_tensor_inputs: typing.List[str] = ['latents'] max_sequence_length: int = 512 ) → ~WanPipelineOutput or tuple

Parameters

prompt (str or List[str], optional) — The prompt or prompts to guide the image generation. If not defined, one has to pass prompt_embeds. instead.
height (int, defaults to 480) — The height in pixels of the generated image.
width (int, defaults to 832) — The width in pixels of the generated image.
num_frames (int, defaults to 81) — The number of frames in the generated video.
num_inference_steps (int, defaults to 50) — The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference.
guidance_scale (float, defaults to 5.0) — Guidance scale as defined in Classifier-Free Diffusion Guidance. guidance_scale is defined as w of equation 2. of Imagen Paper. Guidance scale is enabled by setting guidance_scale > 1. Higher guidance scale encourages to generate images that are closely linked to the text prompt, usually at the expense of lower image quality.
num_videos_per_prompt (int, optional, defaults to 1) — The number of images to generate per prompt.
generator (torch.Generator or List[torch.Generator], optional) — A torch.Generator to make generation deterministic.
latents (torch.Tensor, optional) — Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image generation. Can be used to tweak the same generation with different prompts. If not provided, a latents tensor is generated by sampling using the supplied random generator.
prompt_embeds (torch.Tensor, optional) — Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, text embeddings are generated from the prompt input argument.
output_type (str, optional, defaults to "pil") — The output format of the generated image. Choose between PIL.Image or np.array.
return_dict (bool, optional, defaults to True) — Whether or not to return a WanPipelineOutput instead of a plain tuple.
attention_kwargs (dict, optional) — A kwargs dictionary that if specified is passed along to the AttentionProcessor as defined under self.processor in diffusers.models.attention_processor.
callback_on_step_end (Callable, PipelineCallback, MultiPipelineCallbacks, optional) — A function or a subclass of PipelineCallback or MultiPipelineCallbacks that is called at the end of each denoising step during the inference. with the following arguments: callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict). callback_kwargs will include a list of all tensors as specified by callback_on_step_end_tensor_inputs.
callback_on_step_end_tensor_inputs (List, optional) — The list of tensor inputs for the callback_on_step_end function. The tensors specified in the list will be passed as callback_kwargs argument. You will only be able to include variables listed in the ._callback_tensor_inputs attribute of your pipeline class.
autocast_dtype (torch.dtype, optional, defaults to torch.bfloat16) — The dtype to use for the torch.amp.autocast.

Returns

~WanPipelineOutput or tuple

If return_dict is True, WanPipelineOutput is returned, otherwise a tuple is returned where the first element is a list with the generated images and the second element is a list of bools indicating whether the corresponding generated image contains “not-safe-for-work” (nsfw) content.

The call function to the pipeline for generation.

Examples:

>>> import torch
>>> from diffusers.utils import export_to_video
>>> from diffusers import AutoencoderKLWan, WanPipeline
>>> from diffusers.schedulers.scheduling_unipc_multistep import UniPCMultistepScheduler

>>> # Available models: Wan-AI/Wan2.1-T2V-14B-Diffusers, Wan-AI/Wan2.1-T2V-1.3B-Diffusers
>>> model_id = "Wan-AI/Wan2.1-T2V-14B-Diffusers"
>>> vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)
>>> pipe = WanPipeline.from_pretrained(model_id, vae=vae, torch_dtype=torch.bfloat16)
>>> flow_shift = 5.0  # 5.0 for 720P, 3.0 for 480P
>>> pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config, flow_shift=flow_shift)
>>> pipe.to("cuda")

>>> prompt = "A cat and a dog baking a cake together in a kitchen. The cat is carefully measuring flour, while the dog is stirring the batter with a wooden spoon. The kitchen is cozy, with sunlight streaming through the window."
>>> negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"

>>> output = pipe(
...     prompt=prompt,
...     negative_prompt=negative_prompt,
...     height=720,
...     width=1280,
...     num_frames=81,
...     guidance_scale=5.0,
... ).frames[0]
>>> export_to_video(output, "output.mp4", fps=16)

encode_prompt

< source >

( prompt: typing.Union[str, typing.List[str]] negative_prompt: typing.Union[str, typing.List[str], NoneType] = None do_classifier_free_guidance: bool = True num_videos_per_prompt: int = 1 prompt_embeds: typing.Optional[torch.Tensor] = None negative_prompt_embeds: typing.Optional[torch.Tensor] = None max_sequence_length: int = 226 device: typing.Optional[torch.device] = None dtype: typing.Optional[torch.dtype] = None )

Parameters

prompt (str or List[str], optional) — prompt to be encoded
negative_prompt (str or List[str], optional) — The prompt or prompts not to guide the image generation. If not defined, one has to pass negative_prompt_embeds instead. Ignored when not using guidance (i.e., ignored if guidance_scale is less than 1).
do_classifier_free_guidance (bool, optional, defaults to True) — Whether to use classifier free guidance or not.
num_videos_per_prompt (int, optional, defaults to 1) — Number of videos that should be generated per prompt. torch device to place the resulting embeddings on
prompt_embeds (torch.Tensor, optional) — Pre-generated text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, text embeddings will be generated from prompt input argument.
negative_prompt_embeds (torch.Tensor, optional) — Pre-generated negative text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, negative_prompt_embeds will be generated from negative_prompt input argument.
device — (torch.device, optional): torch device
dtype — (torch.dtype, optional): torch dtype

Encodes the prompt into text encoder hidden states.

WanImageToVideoPipeline

class diffusers.WanImageToVideoPipeline

< source >

( tokenizer: AutoTokenizer text_encoder: UMT5EncoderModel image_encoder: CLIPVisionModel image_processor: CLIPImageProcessor transformer: WanTransformer3DModel vae: AutoencoderKLWan scheduler: FlowMatchEulerDiscreteScheduler )

Parameters

tokenizer (T5Tokenizer) — Tokenizer from T5, specifically the google/umt5-xxl variant.
text_encoder (T5EncoderModel) — T5, specifically the google/umt5-xxl variant.
image_encoder (CLIPVisionModel) — CLIP, specifically the clip-vit-huge-patch14 variant.
transformer (WanTransformer3DModel) — Conditional Transformer to denoise the input latents.
scheduler (UniPCMultistepScheduler) — A scheduler to be used in combination with transformer to denoise the encoded image latents.
vae (AutoencoderKLWan) — Variational Auto-Encoder (VAE) Model to encode and decode videos to and from latent representations.

Pipeline for image-to-video generation using Wan.

This model inherits from DiffusionPipeline. Check the superclass documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular device, etc.).

call

< source >

( image: typing.Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, typing.List[PIL.Image.Image], typing.List[numpy.ndarray], typing.List[torch.Tensor]] prompt: typing.Union[str, typing.List[str]] = None negative_prompt: typing.Union[str, typing.List[str]] = None height: int = 480 width: int = 832 num_frames: int = 81 num_inference_steps: int = 50 guidance_scale: float = 5.0 num_videos_per_prompt: typing.Optional[int] = 1 generator: typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None latents: typing.Optional[torch.Tensor] = None prompt_embeds: typing.Optional[torch.Tensor] = None negative_prompt_embeds: typing.Optional[torch.Tensor] = None output_type: typing.Optional[str] = 'np' return_dict: bool = True attention_kwargs: typing.Optional[typing.Dict[str, typing.Any]] = None callback_on_step_end: typing.Union[typing.Callable[[int, int, typing.Dict], NoneType], diffusers.callbacks.PipelineCallback, diffusers.callbacks.MultiPipelineCallbacks, NoneType] = None callback_on_step_end_tensor_inputs: typing.List[str] = ['latents'] max_sequence_length: int = 512 ) → ~WanPipelineOutput or tuple

Parameters

image (PipelineImageInput) — The input image to condition the generation on. Must be an image, a list of images or a torch.Tensor.
prompt (str or List[str], optional) — The prompt or prompts to guide the image generation. If not defined, one has to pass prompt_embeds. instead.
negative_prompt (str or List[str], optional) — The prompt or prompts not to guide the image generation. If not defined, one has to pass negative_prompt_embeds instead. Ignored when not using guidance (i.e., ignored if guidance_scale is less than 1).
height (int, defaults to 480) — The height of the generated video.
width (int, defaults to 832) — The width of the generated video.
num_frames (int, defaults to 81) — The number of frames in the generated video.
num_inference_steps (int, defaults to 50) — The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference.
guidance_scale (float, defaults to 5.0) — Guidance scale as defined in Classifier-Free Diffusion Guidance. guidance_scale is defined as w of equation 2. of Imagen Paper. Guidance scale is enabled by setting guidance_scale > 1. Higher guidance scale encourages to generate images that are closely linked to the text prompt, usually at the expense of lower image quality.
num_videos_per_prompt (int, optional, defaults to 1) — The number of images to generate per prompt.
generator (torch.Generator or List[torch.Generator], optional) — A torch.Generator to make generation deterministic.
latents (torch.Tensor, optional) — Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image generation. Can be used to tweak the same generation with different prompts. If not provided, a latents tensor is generated by sampling using the supplied random generator.
prompt_embeds (torch.Tensor, optional) — Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, text embeddings are generated from the prompt input argument.
output_type (str, optional, defaults to "pil") — The output format of the generated image. Choose between PIL.Image or np.array.
return_dict (bool, optional, defaults to True) — Whether or not to return a WanPipelineOutput instead of a plain tuple.
attention_kwargs (dict, optional) — A kwargs dictionary that if specified is passed along to the AttentionProcessor as defined under self.processor in diffusers.models.attention_processor.
callback_on_step_end (Callable, PipelineCallback, MultiPipelineCallbacks, optional) — A function or a subclass of PipelineCallback or MultiPipelineCallbacks that is called at the end of each denoising step during the inference. with the following arguments: callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict). callback_kwargs will include a list of all tensors as specified by callback_on_step_end_tensor_inputs.
callback_on_step_end_tensor_inputs (List, optional) — The list of tensor inputs for the callback_on_step_end function. The tensors specified in the list will be passed as callback_kwargs argument. You will only be able to include variables listed in the ._callback_tensor_inputs attribute of your pipeline class.
max_sequence_length (int, optional, defaults to 512) — The maximum sequence length of the prompt.
shift (float, optional, defaults to 5.0) — The shift of the flow.
autocast_dtype (torch.dtype, optional, defaults to torch.bfloat16) — The dtype to use for the torch.amp.autocast.

Returns

~WanPipelineOutput or tuple

The call function to the pipeline for generation.

Examples:

>>> import torch
>>> import numpy as np
>>> from diffusers import AutoencoderKLWan, WanImageToVideoPipeline
>>> from diffusers.utils import export_to_video, load_image
>>> from transformers import CLIPVisionModel

>>> # Available models: Wan-AI/Wan2.1-I2V-14B-480P-Diffusers, Wan-AI/Wan2.1-I2V-14B-720P-Diffusers
>>> model_id = "Wan-AI/Wan2.1-I2V-14B-480P-Diffusers"
>>> image_encoder = CLIPVisionModel.from_pretrained(
...     model_id, subfolder="image_encoder", torch_dtype=torch.float32
... )
>>> vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)
>>> pipe = WanImageToVideoPipeline.from_pretrained(
...     model_id, vae=vae, image_encoder=image_encoder, torch_dtype=torch.bfloat16
... )
>>> pipe.to("cuda")

>>> image = load_image(
...     "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astronaut.jpg"
... )
>>> max_area = 480 * 832
>>> aspect_ratio = image.height / image.width
>>> mod_value = pipe.vae_scale_factor_spatial * pipe.transformer.config.patch_size[1]
>>> height = round(np.sqrt(max_area * aspect_ratio)) // mod_value * mod_value
>>> width = round(np.sqrt(max_area / aspect_ratio)) // mod_value * mod_value
>>> image = image.resize((width, height))
>>> prompt = (
...     "An astronaut hatching from an egg, on the surface of the moon, the darkness and depth of space realised in "
...     "the background. High quality, ultrarealistic detail and breath-taking movie-like camera shot."
... )
>>> negative_prompt = "Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards"

>>> output = pipe(
...     image=image,
...     prompt=prompt,
...     negative_prompt=negative_prompt,
...     height=height,
...     width=width,
...     num_frames=81,
...     guidance_scale=5.0,
... ).frames[0]
>>> export_to_video(output, "output.mp4", fps=16)

encode_prompt

< source >

Parameters

prompt (str or List[str], optional) — prompt to be encoded
negative_prompt (str or List[str], optional) — The prompt or prompts not to guide the image generation. If not defined, one has to pass negative_prompt_embeds instead. Ignored when not using guidance (i.e., ignored if guidance_scale is less than 1).
do_classifier_free_guidance (bool, optional, defaults to True) — Whether to use classifier free guidance or not.
num_videos_per_prompt (int, optional, defaults to 1) — Number of videos that should be generated per prompt. torch device to place the resulting embeddings on
prompt_embeds (torch.Tensor, optional) — Pre-generated text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, text embeddings will be generated from prompt input argument.
negative_prompt_embeds (torch.Tensor, optional) — Pre-generated negative text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, negative_prompt_embeds will be generated from negative_prompt input argument.
device — (torch.device, optional): torch device
dtype — (torch.dtype, optional): torch dtype

Encodes the prompt into text encoder hidden states.

WanPipelineOutput

class diffusers.pipelines.wan.pipeline_output.WanPipelineOutput

< source >

( frames: Tensor )

Parameters

frames (torch.Tensor, np.ndarray, or List[List[PIL.Image.Image]]) — List of video outputs - It can be a nested list of length batch_size, with each sub-list containing denoised PIL image sequences of length num_frames. It can also be a NumPy array or Torch tensor of shape (batch_size, num_frames, channels, height, width).

Output class for Wan pipelines.

< > Update on GitHub

←Value-guided sampling Wuerstchen→

Diffusers

Wan

Generating Videos with Wan 2.1

Text to Video Generation

Image to Video Generation

Video to Video Generation

Memory Optimizations for Wan 2.1

Group Offloading the Transformer and UMT5 Text Encoder

Block Level Group Offloading

Block Level Group Offloading with CUDA Streams

Applying Layerwise Casting to the Transformer

Using a Custom Scheduler

Using Single File Loading with Wan 2.1

Recommendations for Inference

WanPipeline

class diffusers.WanPipeline

__call__

encode_prompt

WanImageToVideoPipeline

class diffusers.WanImageToVideoPipeline

__call__

encode_prompt

WanPipelineOutput

class diffusers.pipelines.wan.pipeline_output.WanPipelineOutput

call

call