update wan training

Revert "Wan refactor"

diffsynth/configs/model_config.py

@@ -133,7 +133,10 @@ model_loader_configs = [
     (None, "3ef3b1f8e1dab83d5b71fd7b617f859f", ["wan_video_dit"], [WanModel], "civitai"),
     (None, "70ddad9d3a133785da5ea371aae09504", ["wan_video_dit"], [WanModel], "civitai"),
     (None, "26bde73488a92e64cc20b0a7485b9e5b", ["wan_video_dit"], [WanModel], "civitai"),
+    (None, "ac6a5aa74f4a0aab6f64eb9a72f19901", ["wan_video_dit"], [WanModel], "civitai"), 
+    (None, "b61c605c2adbd23124d152ed28e049ae", ["wan_video_dit"], [WanModel], "civitai"), 
     (None, "a61453409b67cd3246cf0c3bebad47ba", ["wan_video_dit", "wan_video_vace"], [WanModel, VaceWanModel], "civitai"),
+    (None, "7a513e1f257a861512b1afd387a8ecd9", ["wan_video_dit", "wan_video_vace"], [WanModel, VaceWanModel], "civitai"),
     (None, "cb104773c6c2cb6df4f9529ad5c60d0b", ["wan_video_dit"], [WanModel], "diffusers"),
     (None, "9c8818c2cbea55eca56c7b447df170da", ["wan_video_text_encoder"], [WanTextEncoder], "civitai"),
     (None, "5941c53e207d62f20f9025686193c40b", ["wan_video_image_encoder"], [WanImageEncoder], "civitai"),

diffsynth/models/wan_video_camera_controller.py

@@ -0,0 +1,202 @@
+import torch
+import torch.nn as nn
+import numpy as np
+from einops import rearrange
+import os
+from typing_extensions import Literal
+
+class SimpleAdapter(nn.Module):
+    def __init__(self, in_dim, out_dim, kernel_size, stride, num_residual_blocks=1):
+        super(SimpleAdapter, self).__init__()
+
+        # Pixel Unshuffle: reduce spatial dimensions by a factor of 8
+        self.pixel_unshuffle = nn.PixelUnshuffle(downscale_factor=8)
+
+        # Convolution: reduce spatial dimensions by a factor
+        #  of 2 (without overlap)
+        self.conv = nn.Conv2d(in_dim * 64, out_dim, kernel_size=kernel_size, stride=stride, padding=0)
+
+        # Residual blocks for feature extraction
+        self.residual_blocks = nn.Sequential(
+            *[ResidualBlock(out_dim) for _ in range(num_residual_blocks)]
+        )
+
+    def forward(self, x):
+        # Reshape to merge the frame dimension into batch
+        bs, c, f, h, w = x.size()
+        x = x.permute(0, 2, 1, 3, 4).contiguous().view(bs * f, c, h, w)
+
+        # Pixel Unshuffle operation
+        x_unshuffled = self.pixel_unshuffle(x)
+
+        # Convolution operation
+        x_conv = self.conv(x_unshuffled)
+
+        # Feature extraction with residual blocks
+        out = self.residual_blocks(x_conv)
+
+        # Reshape to restore original bf dimension
+        out = out.view(bs, f, out.size(1), out.size(2), out.size(3))
+
+        # Permute dimensions to reorder (if needed), e.g., swap channels and feature frames
+        out = out.permute(0, 2, 1, 3, 4)
+
+        return out
+    
+    def process_camera_coordinates(
+        self,
+        direction: Literal["Left", "Right", "Up", "Down", "LeftUp", "LeftDown", "RightUp", "RightDown"],
+        length: int,
+        height: int,
+        width: int,
+        speed: float = 1/54,
+        origin=(0, 0.532139961, 0.946026558, 0.5, 0.5, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0)
+    ):
+        if origin is None:
+            origin = (0, 0.532139961, 0.946026558, 0.5, 0.5, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0)
+        coordinates = generate_camera_coordinates(direction, length, speed, origin)
+        plucker_embedding = process_pose_file(coordinates, width, height)
+        return plucker_embedding
+        
+    
+
+class ResidualBlock(nn.Module):
+    def __init__(self, dim):
+        super(ResidualBlock, self).__init__()
+        self.conv1 = nn.Conv2d(dim, dim, kernel_size=3, padding=1)
+        self.relu = nn.ReLU(inplace=True)
+        self.conv2 = nn.Conv2d(dim, dim, kernel_size=3, padding=1)
+
+    def forward(self, x):
+        residual = x
+        out = self.relu(self.conv1(x))
+        out = self.conv2(out)
+        out += residual
+        return out
+    
+class Camera(object):
+    """Copied from https://github.com/hehao13/CameraCtrl/blob/main/inference.py
+    """
+    def __init__(self, entry):
+        fx, fy, cx, cy = entry[1:5]
+        self.fx = fx
+        self.fy = fy
+        self.cx = cx
+        self.cy = cy
+        w2c_mat = np.array(entry[7:]).reshape(3, 4)
+        w2c_mat_4x4 = np.eye(4)
+        w2c_mat_4x4[:3, :] = w2c_mat
+        self.w2c_mat = w2c_mat_4x4
+        self.c2w_mat = np.linalg.inv(w2c_mat_4x4)
+
+def get_relative_pose(cam_params):
+    """Copied from https://github.com/hehao13/CameraCtrl/blob/main/inference.py
+    """
+    abs_w2cs = [cam_param.w2c_mat for cam_param in cam_params]
+    abs_c2ws = [cam_param.c2w_mat for cam_param in cam_params]
+    cam_to_origin = 0
+    target_cam_c2w = np.array([
+        [1, 0, 0, 0],
+        [0, 1, 0, -cam_to_origin],
+        [0, 0, 1, 0],
+        [0, 0, 0, 1]
+    ])
+    abs2rel = target_cam_c2w @ abs_w2cs[0]
+    ret_poses = [target_cam_c2w, ] + [abs2rel @ abs_c2w for abs_c2w in abs_c2ws[1:]]
+    ret_poses = np.array(ret_poses, dtype=np.float32)
+    return ret_poses
+
+def custom_meshgrid(*args):
+    # torch>=2.0.0 only
+    return torch.meshgrid(*args, indexing='ij')
+
+
+def ray_condition(K, c2w, H, W, device):
+    """Copied from https://github.com/hehao13/CameraCtrl/blob/main/inference.py
+    """
+    # c2w: B, V, 4, 4
+    # K: B, V, 4
+
+    B = K.shape[0]
+
+    j, i = custom_meshgrid(
+        torch.linspace(0, H - 1, H, device=device, dtype=c2w.dtype),
+        torch.linspace(0, W - 1, W, device=device, dtype=c2w.dtype),
+    )
+    i = i.reshape([1, 1, H * W]).expand([B, 1, H * W]) + 0.5  # [B, HxW]
+    j = j.reshape([1, 1, H * W]).expand([B, 1, H * W]) + 0.5  # [B, HxW]
+
+    fx, fy, cx, cy = K.chunk(4, dim=-1)  # B,V, 1
+
+    zs = torch.ones_like(i)  # [B, HxW]
+    xs = (i - cx) / fx * zs
+    ys = (j - cy) / fy * zs
+    zs = zs.expand_as(ys)
+
+    directions = torch.stack((xs, ys, zs), dim=-1)  # B, V, HW, 3
+    directions = directions / directions.norm(dim=-1, keepdim=True)  # B, V, HW, 3
+
+    rays_d = directions @ c2w[..., :3, :3].transpose(-1, -2)  # B, V, 3, HW
+    rays_o = c2w[..., :3, 3]  # B, V, 3
+    rays_o = rays_o[:, :, None].expand_as(rays_d)  # B, V, 3, HW
+    # c2w @ dirctions
+    rays_dxo = torch.linalg.cross(rays_o, rays_d)
+    plucker = torch.cat([rays_dxo, rays_d], dim=-1)
+    plucker = plucker.reshape(B, c2w.shape[1], H, W, 6)  # B, V, H, W, 6
+    # plucker = plucker.permute(0, 1, 4, 2, 3)
+    return plucker
+
+
+def process_pose_file(cam_params, width=672, height=384, original_pose_width=1280, original_pose_height=720, device='cpu', return_poses=False):
+    if return_poses:
+        return cam_params
+    else:
+        cam_params = [Camera(cam_param) for cam_param in cam_params]
+
+        sample_wh_ratio = width / height
+        pose_wh_ratio = original_pose_width / original_pose_height  # Assuming placeholder ratios, change as needed
+
+        if pose_wh_ratio > sample_wh_ratio:
+            resized_ori_w = height * pose_wh_ratio
+            for cam_param in cam_params:
+                cam_param.fx = resized_ori_w * cam_param.fx / width
+        else:
+            resized_ori_h = width / pose_wh_ratio
+            for cam_param in cam_params:
+                cam_param.fy = resized_ori_h * cam_param.fy / height
+
+        intrinsic = np.asarray([[cam_param.fx * width,
+                                cam_param.fy * height,
+                                cam_param.cx * width,
+                                cam_param.cy * height]
+                                for cam_param in cam_params], dtype=np.float32)
+
+        K = torch.as_tensor(intrinsic)[None]  # [1, 1, 4]
+        c2ws = get_relative_pose(cam_params)  # Assuming this function is defined elsewhere
+        c2ws = torch.as_tensor(c2ws)[None]  # [1, n_frame, 4, 4]
+        plucker_embedding = ray_condition(K, c2ws, height, width, device=device)[0].permute(0, 3, 1, 2).contiguous()  # V, 6, H, W
+        plucker_embedding = plucker_embedding[None]
+        plucker_embedding = rearrange(plucker_embedding, "b f c h w -> b f h w c")[0]
+        return plucker_embedding
+
+
+
+def generate_camera_coordinates(
+    direction: Literal["Left", "Right", "Up", "Down", "LeftUp", "LeftDown", "RightUp", "RightDown"],
+    length: int,
+    speed: float = 1/54,
+    origin=(0, 0.532139961, 0.946026558, 0.5, 0.5, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0)
+):
+    coordinates = [list(origin)]
+    while len(coordinates) < length:
+        coor = coordinates[-1].copy()
+        if "Left" in direction:
+            coor[9] += speed
+        if "Right" in direction:
+            coor[9] -= speed
+        if "Up" in direction:
+            coor[13] += speed
+        if "Down" in direction:
+            coor[13] -= speed
+        coordinates.append(coor)
+    return coordinates

diffsynth/models/wan_video_dit.py

@@ -5,6 +5,7 @@ import math
 from typing import Tuple, Optional
 from einops import rearrange
 from .utils import hash_state_dict_keys
+from .wan_video_camera_controller import SimpleAdapter
 try:
     import flash_attn_interface
     FLASH_ATTN_3_AVAILABLE = True
@@ -273,6 +274,8 @@ class WanModel(torch.nn.Module):
         has_image_input: bool,
         has_image_pos_emb: bool = False,
         has_ref_conv: bool = False,
+        add_control_adapter: bool = False,
+        in_dim_control_adapter: int = 24,
     ):
         super().__init__()
         self.dim = dim
@@ -308,9 +311,17 @@ class WanModel(torch.nn.Module):
             self.ref_conv = nn.Conv2d(16, dim, kernel_size=(2, 2), stride=(2, 2))
         self.has_image_pos_emb = has_image_pos_emb
         self.has_ref_conv = has_ref_conv
+        if add_control_adapter:
+            self.control_adapter = SimpleAdapter(in_dim_control_adapter, dim, kernel_size=patch_size[1:], stride=patch_size[1:])
+        else:
+            self.control_adapter = None
 
-    def patchify(self, x: torch.Tensor):
+    def patchify(self, x: torch.Tensor,control_camera_latents_input: torch.Tensor = None):
         x = self.patch_embedding(x)
+        if self.control_adapter is not None and control_camera_latents_input is not None:
+            y_camera = self.control_adapter(control_camera_latents_input)
+            x = [u + v for u, v in zip(x, y_camera)]
+            x = x[0].unsqueeze(0)
         grid_size = x.shape[2:]
         x = rearrange(x, 'b c f h w -> b (f h w) c').contiguous()
         return x, grid_size  # x, grid_size: (f, h, w)
@@ -612,6 +623,42 @@ class WanModelStateDictConverter:
                 "eps": 1e-6,
                 "has_ref_conv": True
             }
+        elif hash_state_dict_keys(state_dict) == "ac6a5aa74f4a0aab6f64eb9a72f19901":
+            # 1.3B PAI control-camera v1.1
+            config = {
+                "has_image_input": True,
+                "patch_size": [1, 2, 2],
+                "in_dim": 32,
+                "dim": 1536,
+                "ffn_dim": 8960,
+                "freq_dim": 256,
+                "text_dim": 4096,
+                "out_dim": 16,
+                "num_heads": 12,
+                "num_layers": 30,
+                "eps": 1e-6,
+                "has_ref_conv": False,
+                "add_control_adapter": True,
+                "in_dim_control_adapter": 24,
+            }
+        elif hash_state_dict_keys(state_dict) == "b61c605c2adbd23124d152ed28e049ae":
+            # 14B PAI control-camera v1.1
+            config = {
+                "has_image_input": True,
+                "patch_size": [1, 2, 2],
+                "in_dim": 32,
+                "dim": 5120,
+                "ffn_dim": 13824,
+                "freq_dim": 256,
+                "text_dim": 4096,
+                "out_dim": 16,
+                "num_heads": 40,
+                "num_layers": 40,
+                "eps": 1e-6,
+                "has_ref_conv": False,
+                "add_control_adapter": True,
+                "in_dim_control_adapter": 24,
+            }
         else:
             config = {}
         return state_dict, config

diffsynth/models/wan_video_vace.py

@@ -1,6 +1,6 @@
 import torch
 from .wan_video_dit import DiTBlock
-
+from .utils import hash_state_dict_keys
 
 class VaceWanAttentionBlock(DiTBlock):
     def __init__(self, has_image_input, dim, num_heads, ffn_dim, eps=1e-6, block_id=0):
@@ -50,7 +50,11 @@ class VaceWanModel(torch.nn.Module):
         # vace patch embeddings
         self.vace_patch_embedding = torch.nn.Conv3d(vace_in_dim, dim, kernel_size=patch_size, stride=patch_size)
 
-    def forward(self, x, vace_context, context, t_mod, freqs):
+    def forward(
+        self, x, vace_context, context, t_mod, freqs,
+        use_gradient_checkpointing: bool = False,
+        use_gradient_checkpointing_offload: bool = False,
+    ):
         c = [self.vace_patch_embedding(u.unsqueeze(0)) for u in vace_context]
         c = [u.flatten(2).transpose(1, 2) for u in c]
         c = torch.cat([
@@ -58,8 +62,27 @@ class VaceWanModel(torch.nn.Module):
                       dim=1) for u in c
         ])
         
+        def create_custom_forward(module):
+            def custom_forward(*inputs):
+                return module(*inputs)
+            return custom_forward
+        
         for block in self.vace_blocks:
-            c = block(c, x, context, t_mod, freqs)
+            if use_gradient_checkpointing_offload:
+                with torch.autograd.graph.save_on_cpu():
+                    c = torch.utils.checkpoint.checkpoint(
+                        create_custom_forward(block),
+                        c, x, context, t_mod, freqs,
+                        use_reentrant=False,
+                    )
+            elif use_gradient_checkpointing:
+                c = torch.utils.checkpoint.checkpoint(
+                    create_custom_forward(block),
+                    c, x, context, t_mod, freqs,
+                    use_reentrant=False,
+                )
+            else:
+                c = block(c, x, context, t_mod, freqs)
         hints = torch.unbind(c)[:-1]
         return hints
     
@@ -74,4 +97,17 @@ class VaceWanModelDictConverter:
     
     def from_civitai(self, state_dict):
         state_dict_ = {name: param for name, param in state_dict.items() if name.startswith("vace")}
-        return state_dict_
+        if hash_state_dict_keys(state_dict_) == '3b2726384e4f64837bdf216eea3f310d': # vace 14B
+            config = {
+                "vace_layers": (0, 5, 10, 15, 20, 25, 30, 35),
+                "vace_in_dim": 96,
+                "patch_size": (1, 2, 2),
+                "has_image_input": False,
+                "dim": 5120,
+                "num_heads": 40,
+                "ffn_dim": 13824,
+                "eps": 1e-06,                
+            }
+        else:
+            config = {}
+        return state_dict_, config

diffsynth/pipelines/wan_video_new.py

@@ -1,4 +1,4 @@
-import torch, warnings, glob, os
+import torch, warnings, glob, os, types
 import numpy as np
 from PIL import Image
 from einops import repeat, reduce
@@ -10,6 +10,7 @@ import numpy as np
 from PIL import Image
 from tqdm import tqdm
 from typing import Optional
+from typing_extensions import Literal
 
 from ..models import ModelManager, load_state_dict
 from ..models.wan_video_dit import WanModel, RMSNorm, sinusoidal_embedding_1d
@@ -142,10 +143,8 @@ class BasePipeline(torch.nn.Module):
         self.vram_management_enabled = True
         
         
-    def get_free_vram(self):
+    def get_vram(self):
-        total_memory = torch.cuda.get_device_properties(self.device).total_memory
+        return torch.cuda.mem_get_info(self.device)[1] / (1024 ** 3)
-        allocated_memory = torch.cuda.device_memory_used(self.device)
-        return (total_memory - allocated_memory) / (1024 ** 3)
     
     
     def freeze_except(self, model_names):
@@ -210,8 +209,10 @@ class WanVideoPipeline(BasePipeline):
             WanVideoUnit_ImageEmbedder(),
             WanVideoUnit_FunControl(),
             WanVideoUnit_FunReference(),
+            WanVideoUnit_FunCameraControl(),
             WanVideoUnit_SpeedControl(),
             WanVideoUnit_VACE(),
+            WanVideoUnit_UnifiedSequenceParallel(),
             WanVideoUnit_TeaCache(),
             WanVideoUnit_CfgMerger(),
         ]
@@ -244,7 +245,7 @@ class WanVideoPipeline(BasePipeline):
             vram_limit = None
         else:
             if vram_limit is None:
-                vram_limit = self.get_free_vram()
+                vram_limit = self.get_vram()
             vram_limit = vram_limit - vram_buffer
         if self.text_encoder is not None:
             dtype = next(iter(self.text_encoder.parameters())).dtype
@@ -373,6 +374,30 @@ class WanVideoPipeline(BasePipeline):
                 ),
                 vram_limit=vram_limit,
             )
+            
+            
+    def initialize_usp(self):
+        import torch.distributed as dist
+        from xfuser.core.distributed import initialize_model_parallel, init_distributed_environment
+        dist.init_process_group(backend="nccl", init_method="env://")
+        init_distributed_environment(rank=dist.get_rank(), world_size=dist.get_world_size())
+        initialize_model_parallel(
+            sequence_parallel_degree=dist.get_world_size(),
+            ring_degree=1,
+            ulysses_degree=dist.get_world_size(),
+        )
+        torch.cuda.set_device(dist.get_rank())
+            
+            
+    def enable_usp(self):
+        from xfuser.core.distributed import get_sequence_parallel_world_size
+        from ..distributed.xdit_context_parallel import usp_attn_forward, usp_dit_forward
+
+        for block in self.dit.blocks:
+            block.self_attn.forward = types.MethodType(usp_attn_forward, block.self_attn)
+        self.dit.forward = types.MethodType(usp_dit_forward, self.dit)
+        self.sp_size = get_sequence_parallel_world_size()
+        self.use_unified_sequence_parallel = True
 
 
     @staticmethod
@@ -384,6 +409,7 @@ class WanVideoPipeline(BasePipeline):
         local_model_path: str = "./models",
         skip_download: bool = False,
         redirect_common_files: bool = True,
+        use_usp=False,
     ):
         # Redirect model path
         if redirect_common_files:
@@ -411,6 +437,7 @@ class WanVideoPipeline(BasePipeline):
         
         # Initialize pipeline
         pipe = WanVideoPipeline(device=device, torch_dtype=torch_dtype)
+        if use_usp: pipe.initialize_usp()
         pipe.text_encoder = model_manager.fetch_model("wan_video_text_encoder")
         pipe.dit = model_manager.fetch_model("wan_video_dit")
         pipe.vae = model_manager.fetch_model("wan_video_vae")
@@ -422,6 +449,9 @@ class WanVideoPipeline(BasePipeline):
         tokenizer_config.download_if_necessary(local_model_path, skip_download=skip_download)
         pipe.prompter.fetch_models(pipe.text_encoder)
         pipe.prompter.fetch_tokenizer(tokenizer_config.path)
+        
+        # Unified Sequence Parallel
+        if use_usp: pipe.enable_usp()
         return pipe
 
 
@@ -441,6 +471,10 @@ class WanVideoPipeline(BasePipeline):
         # ControlNet
         control_video: Optional[list[Image.Image]] = None,
         reference_image: Optional[Image.Image] = None,
+        # Camera control
+        camera_control_direction: Optional[Literal["Left", "Right", "Up", "Down", "LeftUp", "LeftDown", "RightUp", "RightDown"]] = None,
+        camera_control_speed: Optional[float] = 1/54,
+        camera_control_origin: Optional[tuple] = (0, 0.532139961, 0.946026558, 0.5, 0.5, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0),
         # VACE
         vace_video: Optional[list[Image.Image]] = None,
         vace_video_mask: Optional[Image.Image] = None,
@@ -480,22 +514,23 @@ class WanVideoPipeline(BasePipeline):
         # Inputs
         inputs_posi = {
             "prompt": prompt,
-            "tea_cache_l1_thresh": tea_cache_l1_thresh, "tea_cache_model_id": tea_cache_model_id,
+            "tea_cache_l1_thresh": tea_cache_l1_thresh, "tea_cache_model_id": tea_cache_model_id, "num_inference_steps": num_inference_steps,
         }
         inputs_nega = {
             "negative_prompt": negative_prompt,
-            "tea_cache_l1_thresh": tea_cache_l1_thresh, "tea_cache_model_id": tea_cache_model_id,
+            "tea_cache_l1_thresh": tea_cache_l1_thresh, "tea_cache_model_id": tea_cache_model_id, "num_inference_steps": num_inference_steps,
         }
         inputs_shared = {
             "input_image": input_image,
             "end_image": end_image,
             "input_video": input_video, "denoising_strength": denoising_strength,
             "control_video": control_video, "reference_image": reference_image,
+            "camera_control_direction": camera_control_direction, "camera_control_speed": camera_control_speed, "camera_control_origin": camera_control_origin,
             "vace_video": vace_video, "vace_video_mask": vace_video_mask, "vace_reference_image": vace_reference_image, "vace_scale": vace_scale,
             "seed": seed, "rand_device": rand_device,
             "height": height, "width": width, "num_frames": num_frames,
             "cfg_scale": cfg_scale, "cfg_merge": cfg_merge,
-            "num_inference_steps": num_inference_steps, "sigma_shift": sigma_shift,
+            "sigma_shift": sigma_shift,
             "motion_bucket_id": motion_bucket_id,
             "tiled": tiled, "tile_size": tile_size, "tile_stride": tile_stride,
             "sliding_window_size": sliding_window_size, "sliding_window_stride": sliding_window_stride,
@@ -616,16 +651,20 @@ class WanVideoUnit_NoiseInitializer(PipelineUnit):
 class WanVideoUnit_InputVideoEmbedder(PipelineUnit):
     def __init__(self):
         super().__init__(
-            input_params=("input_video", "noise", "tiled", "tile_size", "tile_stride", "denoising_strength"),
+            input_params=("input_video", "noise", "tiled", "tile_size", "tile_stride", "vace_reference_image"),
             onload_model_names=("vae",)
         )
 
-    def process(self, pipe: WanVideoPipeline, input_video, noise, tiled, tile_size, tile_stride, denoising_strength):
+    def process(self, pipe: WanVideoPipeline, input_video, noise, tiled, tile_size, tile_stride, vace_reference_image):
         if input_video is None:
             return {"latents": noise}
         pipe.load_models_to_device(["vae"])
         input_video = pipe.preprocess_video(input_video)
         input_latents = pipe.vae.encode(input_video, device=pipe.device, tiled=tiled, tile_size=tile_size, tile_stride=tile_stride).to(dtype=pipe.torch_dtype, device=pipe.device)
+        if vace_reference_image is not None:
+            vace_reference_image = pipe.preprocess_video([vace_reference_image])
+            vace_reference_latents = pipe.vae.encode(vace_reference_image, device=pipe.device).to(dtype=pipe.torch_dtype, device=pipe.device)
+            input_latents = torch.concat([vace_reference_latents, input_latents], dim=2)
         if pipe.scheduler.training:
             return {"latents": noise, "input_latents": input_latents}
         else:
@@ -686,8 +725,8 @@ class WanVideoUnit_ImageEmbedder(PipelineUnit):
         y = y.to(dtype=pipe.torch_dtype, device=pipe.device)
         return {"clip_feature": clip_context, "y": y}
 
-
+        
-
+ 
 class WanVideoUnit_FunControl(PipelineUnit):
     def __init__(self):
         super().__init__(
@@ -732,6 +771,40 @@ class WanVideoUnit_FunReference(PipelineUnit):
 
 
 
+class WanVideoUnit_FunCameraControl(PipelineUnit):
+    def __init__(self):
+        super().__init__(
+            input_params=("height", "width", "num_frames", "camera_control_direction", "camera_control_speed", "camera_control_origin", "latents", "input_image")
+        )
+
+    def process(self, pipe: WanVideoPipeline, height, width, num_frames, camera_control_direction, camera_control_speed, camera_control_origin, latents, input_image):
+        if camera_control_direction is None:
+            return {}
+        camera_control_plucker_embedding = pipe.dit.control_adapter.process_camera_coordinates(
+            camera_control_direction, num_frames, height, width, camera_control_speed, camera_control_origin)
+        
+        control_camera_video = camera_control_plucker_embedding[:num_frames].permute([3, 0, 1, 2]).unsqueeze(0)
+        control_camera_latents = torch.concat(
+            [
+                torch.repeat_interleave(control_camera_video[:, :, 0:1], repeats=4, dim=2),
+                control_camera_video[:, :, 1:]
+            ], dim=2
+        ).transpose(1, 2)
+        b, f, c, h, w = control_camera_latents.shape
+        control_camera_latents = control_camera_latents.contiguous().view(b, f // 4, 4, c, h, w).transpose(2, 3)
+        control_camera_latents = control_camera_latents.contiguous().view(b, f // 4, c * 4, h, w).transpose(1, 2)
+        control_camera_latents_input = control_camera_latents.to(device=pipe.device, dtype=pipe.torch_dtype)
+
+        input_image = input_image.resize((width, height))
+        input_latents = pipe.preprocess_video([input_image])
+        input_latents = pipe.vae.encode(input_latents, device=pipe.device)
+        y = torch.zeros_like(latents).to(pipe.device)
+        y[:, :, :1] = input_latents
+        y = y.to(dtype=pipe.torch_dtype, device=pipe.device)
+        return {"control_camera_latents_input": control_camera_latents_input, "y": y}
+
+
+
 class WanVideoUnit_SpeedControl(PipelineUnit):
     def __init__(self):
         super().__init__(input_params=("motion_bucket_id",))
@@ -795,6 +868,18 @@ class WanVideoUnit_VACE(PipelineUnit):
 
 
 
+class WanVideoUnit_UnifiedSequenceParallel(PipelineUnit):
+    def __init__(self):
+        super().__init__(input_params=())
+
+    def process(self, pipe: WanVideoPipeline):
+        if hasattr(pipe, "use_unified_sequence_parallel"):
+            if pipe.use_unified_sequence_parallel:
+                return {"use_unified_sequence_parallel": True}
+        return {}
+
+
+
 class WanVideoUnit_TeaCache(PipelineUnit):
     def __init__(self):
         super().__init__(
@@ -954,6 +1039,7 @@ def model_fn_wan_video(
     cfg_merge: bool = False,
     use_gradient_checkpointing: bool = False,
     use_gradient_checkpointing_offload: bool = False,
+    control_camera_latents_input = None,
     **kwargs,
 ):
     if sliding_window_size is not None and sliding_window_stride is not None:
@@ -1006,7 +1092,9 @@ def model_fn_wan_video(
         clip_embdding = dit.img_emb(clip_feature)
         context = torch.cat([clip_embdding, context], dim=1)
     
-    x, (f, h, w) = dit.patchify(x)
+    # Add camera control
+    x, (f, h, w) = dit.patchify(x, control_camera_latents_input)
+
     
     # Reference image
     if reference_latents is not None:

diffsynth/trainers/utils.py

@@ -11,8 +11,9 @@ class VideoDataset(torch.utils.data.Dataset):
     def __init__(
         self,
         base_path=None, metadata_path=None,
-        frame_interval=1, num_frames=81,
+        num_frames=81,
-        dynamic_resolution=True, max_pixels=1920*1080, height=None, width=None,
+        time_division_factor=4, time_division_remainder=1,
+        max_pixels=1920*1080, height=None, width=None,
         height_division_factor=16, width_division_factor=16,
         data_file_keys=("video",),
         image_file_extension=("jpg", "jpeg", "png", "webp"),
@@ -25,17 +26,15 @@ class VideoDataset(torch.utils.data.Dataset):
             metadata_path = args.dataset_metadata_path
             height = args.height
             width = args.width
+            max_pixels = args.max_pixels
             num_frames = args.num_frames
             data_file_keys = args.data_file_keys.split(",")
             repeat = args.dataset_repeat
-
-        metadata = pd.read_csv(metadata_path)
-        self.data = [metadata.iloc[i].to_dict() for i in range(len(metadata))]
         
         self.base_path = base_path
-        self.frame_interval = frame_interval
         self.num_frames = num_frames
-        self.dynamic_resolution = dynamic_resolution
+        self.time_division_factor = time_division_factor
+        self.time_division_remainder = time_division_remainder
         self.max_pixels = max_pixels
         self.height = height
         self.width = width
@@ -46,9 +45,43 @@ class VideoDataset(torch.utils.data.Dataset):
         self.video_file_extension = video_file_extension
         self.repeat = repeat
         
-        if height is not None and width is not None and dynamic_resolution == True:
+        if height is not None and width is not None:
             print("Height and width are fixed. Setting `dynamic_resolution` to False.")
             self.dynamic_resolution = False
+        elif height is None and width is None:
+            print("Height and width are none. Setting `dynamic_resolution` to True.")
+            self.dynamic_resolution = True
+            
+        if metadata_path is None:
+            print("No metadata. Trying to generate it.")
+            metadata = self.generate_metadata(base_path)
+            print(f"{len(metadata)} lines in metadata.")
+        else:
+            metadata = pd.read_csv(metadata_path)
+        self.data = [metadata.iloc[i].to_dict() for i in range(len(metadata))]
+            
+    
+    def generate_metadata(self, folder):
+        video_list, prompt_list = [], []
+        file_set = set(os.listdir(folder))
+        for file_name in file_set:
+            if "." not in file_name:
+                continue
+            file_ext_name = file_name.split(".")[-1].lower()
+            file_base_name = file_name[:-len(file_ext_name)-1]
+            if file_ext_name not in self.image_file_extension and file_ext_name not in self.video_file_extension:
+                continue
+            prompt_file_name = file_base_name + ".txt"
+            if prompt_file_name not in file_set:
+                continue
+            with open(os.path.join(folder, prompt_file_name), "r", encoding="utf-8") as f:
+                prompt = f.read().strip()
+            video_list.append(file_name)
+            prompt_list.append(prompt)
+        metadata = pd.DataFrame()
+        metadata["video"] = video_list
+        metadata["prompt"] = prompt_list
+        return metadata
         
         
     def crop_and_resize(self, image, target_height, target_width):
@@ -75,15 +108,22 @@ class VideoDataset(torch.utils.data.Dataset):
             height, width = self.height, self.width
         return height, width
     
+    
+    def get_num_frames(self, reader):
+        num_frames = self.num_frames
+        if int(reader.count_frames()) < num_frames:
+            num_frames = int(reader.count_frames())
+            while num_frames > 1 and num_frames % self.time_division_factor != self.time_division_remainder:
+                num_frames -= 1
+        return num_frames
+    
 
-    def load_frames_using_imageio(self, file_path, start_frame_id, interval, num_frames):
+    def load_video(self, file_path):
         reader = imageio.get_reader(file_path)
-        if reader.count_frames() - 1 < start_frame_id + (num_frames - 1) * interval:
+        num_frames = self.get_num_frames(reader)
-            reader.close()
-            return None
         frames = []
         for frame_id in range(num_frames):
-            frame = reader.get_data(start_frame_id + frame_id * interval)
+            frame = reader.get_data(frame_id)
             frame = Image.fromarray(frame)
             frame = self.crop_and_resize(frame, *self.get_height_width(frame))
             frames.append(frame)
@@ -95,11 +135,6 @@ class VideoDataset(torch.utils.data.Dataset):
         image = Image.open(file_path).convert("RGB")
         image = self.crop_and_resize(image, *self.get_height_width(image))
         return image
-
-
-    def load_video(self, file_path):
-        frames = self.load_frames_using_imageio(file_path, 0, self.frame_interval, self.num_frames)
-        return frames
     
     
     def is_image(self, file_path):
@@ -182,34 +217,50 @@ class DiffusionTrainingModule(torch.nn.Module):
 
 
 
-def launch_training_task(model: DiffusionTrainingModule, dataset, learning_rate=1e-4, num_epochs=1, output_path="./models", remove_prefix_in_ckpt=None, args=None):
+class ModelLogger:
-    if args is not None:
+    def __init__(self, output_path, remove_prefix_in_ckpt=None):
-        learning_rate = args.learning_rate
+        self.output_path = output_path
-        num_epochs = args.num_epochs
+        self.remove_prefix_in_ckpt = remove_prefix_in_ckpt
-        output_path = args.output_path
+        
-        remove_prefix_in_ckpt = args.remove_prefix_in_ckpt
-    dataloader = torch.utils.data.DataLoader(dataset, shuffle=True, collate_fn=lambda x: x[0])
-    optimizer = torch.optim.AdamW(model.trainable_modules(), lr=learning_rate)
-    scheduler = torch.optim.lr_scheduler.ConstantLR(optimizer)
     
-    accelerator = Accelerator(gradient_accumulation_steps=1)
+    def on_step_end(self, loss):
+        pass
+    
+    
+    def on_epoch_end(self, accelerator, model, epoch_id):
+        accelerator.wait_for_everyone()
+        if accelerator.is_main_process:
+            state_dict = accelerator.get_state_dict(model)
+            state_dict = accelerator.unwrap_model(model).export_trainable_state_dict(state_dict, remove_prefix=self.remove_prefix_in_ckpt)
+            os.makedirs(self.output_path, exist_ok=True)
+            path = os.path.join(self.output_path, f"epoch-{epoch_id}.safetensors")
+            accelerator.save(state_dict, path, safe_serialization=True)
+
+
+
+def launch_training_task(
+    dataset: torch.utils.data.Dataset,
+    model: DiffusionTrainingModule,
+    model_logger: ModelLogger,
+    optimizer: torch.optim.Optimizer,
+    scheduler: torch.optim.lr_scheduler.LRScheduler,
+    num_epochs: int = 1,
+    gradient_accumulation_steps: int = 1,
+):
+    dataloader = torch.utils.data.DataLoader(dataset, shuffle=True, collate_fn=lambda x: x[0])
+    accelerator = Accelerator(gradient_accumulation_steps=gradient_accumulation_steps)
     model, optimizer, dataloader, scheduler = accelerator.prepare(model, optimizer, dataloader, scheduler)
     
-    for epoch in range(num_epochs):
+    for epoch_id in range(num_epochs):
         for data in tqdm(dataloader):
             with accelerator.accumulate(model):
                 optimizer.zero_grad()
                 loss = model(data)
                 accelerator.backward(loss)
                 optimizer.step()
-        scheduler.step()
+                model_logger.on_step_end(loss)
-        accelerator.wait_for_everyone()
+                scheduler.step()
-        if accelerator.is_main_process:
+        model_logger.on_epoch_end(accelerator, model, epoch_id)
-            state_dict = accelerator.get_state_dict(model)
-            state_dict = accelerator.unwrap_model(model).export_trainable_state_dict(state_dict, remove_prefix=remove_prefix_in_ckpt)
-            os.makedirs(output_path, exist_ok=True)
-            path = os.path.join(output_path, f"epoch-{epoch}.safetensors")
-            accelerator.save(state_dict, path, safe_serialization=True)
 
 
 
@@ -228,30 +279,26 @@ def launch_data_process_task(model: DiffusionTrainingModule, dataset, output_pat
 
 def wan_parser():
     parser = argparse.ArgumentParser(description="Simple example of a training script.")
-    parser.add_argument("--dataset_base_path", type=str, default="", help="Base path of the Dataset.")
+    parser.add_argument("--dataset_base_path", type=str, default="", required=True, help="Base path of the dataset.")
-    parser.add_argument("--dataset_metadata_path", type=str, default="", required=True, help="Metadata path of the Dataset.")
+    parser.add_argument("--dataset_metadata_path", type=str, default=None, help="Path to the metadata file of the dataset.")
-    parser.add_argument("--height", type=int, default=None, help="Image or video height. Leave `height` and `width` None to enable dynamic resolution.")
+    parser.add_argument("--max_pixels", type=int, default=1280*720, help="Maximum number of pixels per frame, used for dynamic resolution..")
-    parser.add_argument("--width", type=int, default=None, help="Image or video width. Leave `height` and `width` None to enable dynamic resolution.")
+    parser.add_argument("--height", type=int, default=None, help="Height of images or videos. Leave `height` and `width` empty to enable dynamic resolution.")
-    parser.add_argument("--num_frames", type=int, default=81, help="Number of frames in each video. The frames are sampled from the prefix.")
+    parser.add_argument("--width", type=int, default=None, help="Width of images or videos. Leave `height` and `width` empty to enable dynamic resolution.")
-    parser.add_argument("--data_file_keys", type=str, default="image,video", help="Data file keys in metadata. Separated by commas.")
+    parser.add_argument("--num_frames", type=int, default=81, help="Number of frames per video. Frames are sampled from the video prefix.")
-    parser.add_argument("--dataset_repeat", type=int, default=1, help="Number of times the dataset is repeated in each epoch.")
+    parser.add_argument("--data_file_keys", type=str, default="image,video", help="Data file keys in the metadata. Comma-separated.")
-    parser.add_argument("--model_paths", type=str, default=None, help="Model paths to be loaded. JSON format.")
+    parser.add_argument("--dataset_repeat", type=int, default=1, help="Number of times to repeat the dataset per epoch.")
-    parser.add_argument("--model_id_with_origin_paths", type=str, default=None, help="Model ID with origin path, e.g., Wan-AI/Wan2.1-T2V-1.3B:diffusion_pytorch_model*.safetensors. Separated by commas.")
+    parser.add_argument("--model_paths", type=str, default=None, help="Paths to load models. In JSON format.")
+    parser.add_argument("--model_id_with_origin_paths", type=str, default=None, help="Model ID with origin paths, e.g., Wan-AI/Wan2.1-T2V-1.3B:diffusion_pytorch_model*.safetensors. Comma-separated.")
     parser.add_argument("--learning_rate", type=float, default=1e-4, help="Learning rate.")
     parser.add_argument("--num_epochs", type=int, default=1, help="Number of epochs.")
-    parser.add_argument("--output_path", type=str, default="./models", help="Save path.")
+    parser.add_argument("--output_path", type=str, default="./models", help="Output save path.")
     parser.add_argument("--remove_prefix_in_ckpt", type=str, default="pipe.dit.", help="Remove prefix in ckpt.")
-    parser.add_argument("--trainable_models", type=str, default=None, help="Trainable models, e.g., dit, vae, text_encoder.")
+    parser.add_argument("--trainable_models", type=str, default=None, help="Models to train, e.g., dit, vae, text_encoder.")
-    parser.add_argument("--lora_base_model", type=str, default=None, help="Add LoRA on which model.")
+    parser.add_argument("--lora_base_model", type=str, default=None, help="Which model LoRA is added to.")
-    parser.add_argument("--lora_target_modules", type=str, default="q,k,v,o,ffn.0,ffn.2", help="Add LoRA on which layer.")
+    parser.add_argument("--lora_target_modules", type=str, default="q,k,v,o,ffn.0,ffn.2", help="Which layers LoRA is added to.")
-    parser.add_argument("--lora_rank", type=int, default=32, help="LoRA rank.")
+    parser.add_argument("--lora_rank", type=int, default=32, help="Rank of LoRA.")
-    parser.add_argument("--input_contains_input_image", default=False, action="store_true", help="Model input contains 'input_image'.")
+    parser.add_argument("--extra_inputs", default=None, help="Additional model inputs, comma-separated.")
-    parser.add_argument("--input_contains_end_image", default=False, action="store_true", help="Model input contains 'end_image'.")
+    parser.add_argument("--use_gradient_checkpointing_offload", default=False, action="store_true", help="Whether to offload gradient checkpointing to CPU memory.")
-    parser.add_argument("--input_contains_control_video", default=False, action="store_true", help="Model input contains 'control_video'.")
+    parser.add_argument("--gradient_accumulation_steps", type=int, default=1, help="Gradient accumulation steps.")
-    parser.add_argument("--input_contains_reference_image", default=False, action="store_true", help="Model input contains 'reference_image'.")
-    parser.add_argument("--input_contains_vace_video", default=False, action="store_true", help="Model input contains 'vace_video'.")
-    parser.add_argument("--input_contains_vace_reference_image", default=False, action="store_true", help="Model input contains 'vace_reference_image'.")
-    parser.add_argument("--input_contains_motion_bucket_id", default=False, action="store_true", help="Model input contains 'motion_bucket_id'.")
-    parser.add_argument("--use_gradient_checkpointing_offload", default=False, action="store_true", help="Offload gradient checkpointing to RAM.")
     return parser
 

diffsynth/vram_management/layers.py

@@ -13,7 +13,8 @@ class AutoTorchModule(torch.nn.Module):
         super().__init__()
         
     def check_free_vram(self):
-        used_memory = torch.cuda.device_memory_used(self.computation_device) / (1024 ** 3)
+        gpu_mem_state = torch.cuda.mem_get_info(self.computation_device)
+        used_memory = (gpu_mem_state[1] - gpu_mem_state[0]) / (1024 ** 3)
         return used_memory < self.vram_limit
 
     def offload(self):

examples/wanvideo/README.md

@@ -1,32 +1,417 @@
+# Wan 2.1
+
+[切换到中文](./README_zh.md)
+
+Wan 2.1 is a collection of video synthesis models open-sourced by Alibaba.
+
+**DiffSynth-Studio has adopted a new inference and training framework. To use the previous version, please click [here](https://github.com/modelscope/DiffSynth-Studio/tree/3edf3583b1f08944cee837b94d9f84d669c2729c).**
+
+## Installation
+
+Before using this model, please install DiffSynth-Studio from **source code**.
+
+```shell
+git clone https://github.com/modelscope/DiffSynth-Studio.git
+cd DiffSynth-Studio
+pip install -e .
+```
+
+## Overview
+
+| Model ID | Extra Parameters | Inference | Full Training | Full Training Validation | LoRA Training | LoRA Training Validation |
+|-|-|-|-|-|-|-|
+|[Wan-AI/Wan2.1-T2V-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-1.3B)||[code](./model_inference/Wan2.1-T2V-1.3B.py)|[code](./model_training/full/Wan2.1-T2V-1.3B.sh)|[code](./model_training/validate_full/Wan2.1-T2V-1.3B.py)|[code](./model_training/lora/Wan2.1-T2V-1.3B.sh)|[code](./model_training/validate_lora/Wan2.1-T2V-1.3B.py)|
+|[Wan-AI/Wan2.1-T2V-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-14B)||[code](./model_inference/Wan2.1-T2V-14B.py)|[code](./model_training/full/Wan2.1-T2V-14B.sh)|[code](./model_training/validate_full/Wan2.1-T2V-14B.py)|[code](./model_training/lora/Wan2.1-T2V-14B.sh)|[code](./model_training/validate_lora/Wan2.1-T2V-14B.py)|
+|[Wan-AI/Wan2.1-I2V-14B-480P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P)|`input_image`|[code](./model_inference/Wan2.1-I2V-14B-480P.py)|[code](./model_training/full/Wan2.1-I2V-14B-480P.sh)|[code](./model_training/validate_full/Wan2.1-I2V-14B-480P.py)|[code](./model_training/lora/Wan2.1-I2V-14B-480P.sh)|[code](./model_training/validate_lora/Wan2.1-I2V-14B-480P.py)|
+|[Wan-AI/Wan2.1-I2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-720P)|`input_image`|[code](./model_inference/Wan2.1-I2V-14B-720P.py)|[code](./model_training/full/Wan2.1-I2V-14B-720P.sh)|[code](./model_training/validate_full/Wan2.1-I2V-14B-720P.py)|[code](./model_training/lora/Wan2.1-I2V-14B-720P.sh)|[code](./model_training/validate_lora/Wan2.1-I2V-14B-720P.py)|
+|[Wan-AI/Wan2.1-FLF2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-FLF2V-14B-720P)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-FLF2V-14B-720P.py)|[code](./model_training/full/Wan2.1-FLF2V-14B-720P.sh)|[code](./model_training/validate_full/Wan2.1-FLF2V-14B-720P.py)|[code](./model_training/lora/Wan2.1-FLF2V-14B-720P.sh)|[code](./model_training/validate_lora/Wan2.1-FLF2V-14B-720P.py)|
+|[PAI/Wan2.1-Fun-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-1.3B-InP.py)|[code](./model_training/full/Wan2.1-Fun-1.3B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-1.3B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-1.3B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)|`control_video`|[code](./model_inference/Wan2.1-Fun-1.3B-Control.py)|[code](./model_training/full/Wan2.1-Fun-1.3B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-1.3B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-1.3B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-14B-InP.py)|[code](./model_training/full/Wan2.1-Fun-14B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-14B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-14B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-14B-InP.py)|
+|[PAI/Wan2.1-Fun-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-Control)|`control_video`|[code](./model_inference/Wan2.1-Fun-14B-Control.py)|[code](./model_training/full/Wan2.1-Fun-14B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-14B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-14B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)|`control_video`, `reference_image`|[code](./model_inference/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control)|`control_video`, `reference_image`|[code](./model_inference/Wan2.1-Fun-V1.1-14B-Control.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-V1.1-14B-InP.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-14B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-14B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera)|`control_camera_video`, `input_image`|[code](./model_inference/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)|`control_camera_video`, `input_image`|[code](./model_inference/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|
+|[iic/VACE-Wan2.1-1.3B-Preview](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)|`vace_control_video`, `vace_reference_image`|[code](./model_inference/Wan2.1-VACE-1.3B-Preview.py)|[code](./model_training/full/Wan2.1-VACE-1.3B-Preview.sh)|[code](./model_training/validate_full/Wan2.1-VACE-1.3B-Preview.py)|[code](./model_training/lora/Wan2.1-VACE-1.3B-Preview.sh)|[code](./model_training/validate_lora/Wan2.1-VACE-1.3B-Preview.py)|
+|[Wan-AI/Wan2.1-VACE-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-1.3B)|`vace_control_video`, `vace_reference_image`|[code](./model_inference/Wan2.1-VACE-1.3B.py)|[code](./model_training/full/Wan2.1-VACE-1.3B.sh)|[code](./model_training/validate_full/Wan2.1-VACE-1.3B.py)|[code](./model_training/lora/Wan2.1-VACE-1.3B.sh)|[code](./model_training/validate_lora/Wan2.1-VACE-1.3B.py)|
+|[Wan-AI/Wan2.1-VACE-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-14B)|`vace_control_video`, `vace_reference_image`|[code](./model_inference/Wan2.1-VACE-14B.py)|[code](./model_training/full/Wan2.1-VACE-14B.sh)|[code](./model_training/validate_full/Wan2.1-VACE-14B.py)|[code](./model_training/lora/Wan2.1-VACE-14B.sh)|[code](./model_training/validate_lora/Wan2.1-VACE-14B.py)|
+|[DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1)|`motion_bucket_id`|[code](./model_inference/Wan2.1-1.3b-speedcontrol-v1.py)|[code](./model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](./model_training/validate_full/Wan2.1-1.3b-speedcontrol-v1.py)|[code](./model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](./model_training/validate_lora/Wan2.1-1.3b-speedcontrol-v1.py)|
 
 
-* dataset
+## Model Inference
-  * `--dataset_base_path`: Base path of the Dataset.
+
-  * `--dataset_metadata_path`: Metadata path of the Dataset.
+The following sections will help you understand our functionalities and write inference code.
-  * `--height`: Image or video height. Leave `height` and `width` None to enable dynamic resolution.
+
-  * `--width`: Image or video width. Leave `height` and `width` None to enable dynamic resolution.
+<details>
-  * `--num_frames`: Number of frames in each video. The frames are sampled from the prefix.
+
-  * `--data_file_keys`: Data file keys in metadata. Separated by commas.
+<summary>Loading the Model</summary>
-  * `--dataset_repeat`: Number of times the dataset is repeated in each epoch.
+
-* Model
+The model is loaded using `from_pretrained`:
-  * `--model_paths`: Model paths to be loaded. JSON format.
+
-  * `--model_id_with_origin_paths`: Model ID with original path, e.g., Wan-AI/Wan2.1-T2V-1.3B:diffusion_pytorch_model*.safetensors. Separated by commas.
+```python
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="Wan2.1_VAE.pth"),
+    ],
+)
+```
+
+Here, `torch_dtype` and `device` specify the computation precision and device respectively. The `model_configs` can be used to configure model paths in various ways:
+
+* Downloading the model from [ModelScope](https://modelscope.cn/) and loading it. In this case, both `model_id` and `origin_file_pattern` need to be specified, for example:
+
+```python
+ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors")
+```
+
+* Loading the model from a local file path. In this case, the `path` parameter needs to be specified, for example:
+
+```python
+ModelConfig(path="models/Wan-AI/Wan2.1-T2V-1.3B/diffusion_pytorch_model.safetensors")
+```
+
+For models that are loaded from multiple files, simply use a list, for example:
+
+```python
+ModelConfig(path=[
+    "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00001-of-00006.safetensors",
+    "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00002-of-00006.safetensors",
+    "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00003-of-00006.safetensors",
+    "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00004-of-00006.safetensors",
+    "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00005-of-00006.safetensors",
+    "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00006-of-00006.safetensors",
+])
+```
+
+The `from_pretrained` function also provides additional parameters to control the behavior during model loading:
+
+* `tokenizer_config`: Path to the tokenizer of the Wan model. Default value is `ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="google/*")`.
+* `local_model_path`: Path where downloaded models are saved. Default value is `"./models"`.
+* `skip_download`: Whether to skip downloading models. Default value is `False`. When your network cannot access [ModelScope](https://modelscope.cn/), manually download the necessary files and set this to `True`.
+* `redirect_common_files`: Whether to redirect duplicate model files. Default value is `True`. Since the Wan series models include multiple base models, some modules like text encoder are shared across these models. To avoid redundant downloads, we redirect the model paths.
+* `use_usp`: Whether to enable Unified Sequence Parallel. Default value is `False`. Used for multi-GPU parallel inference.
+
+</details>
+
+<details>
+
+<summary>VRAM Management</summary>
+
+DiffSynth-Studio provides fine-grained VRAM management for the Wan model, allowing it to run on devices with limited VRAM. You can enable offloading functionality via the following code, which moves parts of the model to system memory on devices with limited VRAM:
+
+```python
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+pipe.enable_vram_management()
+```
+
+FP8 quantization is also supported:
+
+```python
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="Wan2.1_VAE.pth", offload_dtype=torch.float8_e4m3fn),
+    ],
+)
+pipe.enable_vram_management()
+```
+
+Both FP8 quantization and offloading can be enabled simultaneously:
+
+```python
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+    ],
+)
+pipe.enable_vram_management()
+```
+
+FP8 quantization significantly reduces VRAM usage but does not accelerate computations. Some models may experience issues such as blurry, torn, or distorted outputs due to insufficient precision when using FP8 quantization. Use FP8 quantization with caution.
+
+The `enable_vram_management` function provides the following parameters to control VRAM usage:
+
+* `vram_limit`: VRAM usage limit (in GB). By default, it uses all available VRAM on the device. Note that this is not an absolute limit; if the specified VRAM is insufficient but more VRAM is actually available, inference will proceed using the minimum required VRAM.
+* `vram_buffer`: Size of the VRAM buffer (in GB). Default is 0.5GB. Since certain large neural network layers may consume more VRAM unpredictably during their execution phase, a VRAM buffer is necessary. Ideally, this should match the maximum VRAM consumed by any single layer in the model.
+* `num_persistent_param_in_dit`: Number of persistent parameters in DiT models. By default, there is no limit. We plan to remove this parameter in the future, so please avoid relying on it.
+
+</details>
+
+<details>
+
+<summary>Inference Acceleration</summary>
+
+Wan supports multiple acceleration techniques, including:
+
+* **Efficient attention implementations**: If any of these attention implementations are installed in your Python environment, they will be automatically enabled in the following priority:
+    * [Flash Attention 3](https://github.com/Dao-AILab/flash-attention)  
+    * [Flash Attention 2](https://github.com/Dao-AILab/flash-attention)  
+    * [Sage Attention](https://github.com/thu-ml/SageAttention)  
+    * [torch SDPA](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html)   (default setting; we recommend installing `torch>=2.5.0`)
+* **Unified Sequence Parallel**: Sequence parallelism based on [xDiT](https://github.com/xdit-project/xDiT). Please refer to [this example](./acceleration/unified_sequence_parallel.py), and run it using the command: 
+
+```shell
+pip install xfuser>=0.4.3
+torchrun --standalone --nproc_per_node=8 examples/wanvideo/acceleration/unified_sequence_parallel.py
+```
+
+* **TeaCache**: Acceleration technique [TeaCache](https://github.com/ali-vilab/TeaCache). Please refer to [this example](./acceleration/teacache.py).
+
+</details>
+
+
+<details>
+
+<summary>Input Parameters</summary>
+
+The pipeline accepts the following input parameters during inference:
+
+* `prompt`: Prompt describing the content to appear in the video.
+* `negative_prompt`: Negative prompt describing content that should not appear in the video. Default is `""`.
+* `input_image`: Input image, applicable for image-to-video models such as [`Wan-AI/Wan2.1-I2V-14B-480P`](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P) and [`PAI/Wan2.1-Fun-1.3B-InP`](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP), as well as first-and-last-frame models like [`Wan-AI/Wan2.1-FLF2V-14B-720P`](Wan-AI/Wan2.1-FLF2V-14B-720P).
+* `end_image`: End frame, applicable for first-and-last-frame models such as [`Wan-AI/Wan2.1-FLF2V-14B-720P`](Wan-AI/Wan2.1-FLF2V-14B-720P).
+* `input_video`: Input video used for video-to-video generation. Applicable to any Wan series model and must be used together with `denoising_strength`.
+* `denoising_strength`: Denoising strength in range [0, 1]. A smaller value results in a video closer to `input_video`.
+* `control_video`: Control video, applicable to Wan models with control capabilities such as [`PAI/Wan2.1-Fun-1.3B-Control`](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control).  
+* `reference_image`: Reference image, applicable to Wan models supporting reference images such as [`PAI/Wan2.1-Fun-V1.1-1.3B-Control`](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control).  
+* `camera_control_direction`: Camera control direction, optional values are "Left", "Right", "Up", "Down", "LeftUp", "LeftDown", "RightUp", "RightDown". Applicable to Camera-Control models, such as [PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://www.modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera).  
+* `camera_control_speed`: Camera control speed. Applicable to Camera-Control models, such as [PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://www.modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera).  
+* `camera_control_origin`: Origin coordinate of the camera control sequence. Please refer to the [original paper](https://arxiv.org/pdf/2404.02101) for proper configuration. Applicable to Camera-Control models, such as [PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://www.modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera).
+* `vace_video`: Input video for VACE models, applicable to the VACE series such as [`iic/VACE-Wan2.1-1.3B-Preview`](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview).  
+* `vace_video_mask`: Mask video for VACE models, applicable to the VACE series such as [`iic/VACE-Wan2.1-1.3B-Preview`](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview).  
+* `vace_reference_image`: Reference image for VACE models, applicable to the VACE series such as [`iic/VACE-Wan2.1-1.3B-Preview`](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview).  
+* `vace_scale`: Influence of the VACE model on the base model, default is 1. Higher values increase control strength but may lead to visual artifacts or breakdowns.
+* `seed`: Random seed. Default is `None`, meaning fully random.
+* `rand_device`: Device used to generate random Gaussian noise matrix. Default is `"cpu"`. When set to `"cuda"`, different GPUs may produce different generation results.
+* `height`: Frame height, default is 480. Must be a multiple of 16; if not, it will be rounded up.
+* `width`: Frame width, default is 832. Must be a multiple of 16; if not, it will be rounded up.
+* `num_frames`: Number of frames, default is 81. Must be a multiple of 4 plus 1; if not, it will be rounded up, minimum is 1.
+* `cfg_scale`: Classifier-free guidance scale, default is 5. Higher values increase adherence to the prompt but may cause visual artifacts.
+* `cfg_merge`: Whether to merge both sides of classifier-free guidance for unified inference. Default is `False`. This parameter currently only works for basic text-to-video and image-to-video models.
+* `num_inference_steps`: Number of inference steps, default is 50.
+* `sigma_shift`: Parameter from Rectified Flow theory, default is 5. Higher values make the model stay longer at the initial denoising stage. Increasing this may improve video quality but may also cause inconsistency between generated videos and training data due to deviation from training behavior.
+* `motion_bucket_id`: Motion intensity, range [0, 100], applicable to motion control modules such as [`DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1`](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1). Larger values indicate more intense motion.  
+* `tiled`: Whether to enable tiled VAE inference, default is `False`. Setting to `True` significantly reduces VRAM usage during VAE encoding/decoding but introduces small errors and slightly increases inference time.
+* `tile_size`: Tile size during VAE encoding/decoding, default is (30, 52), only effective when `tiled=True`.
+* `tile_stride`: Stride of tiles during VAE encoding/decoding, default is (15, 26), only effective when `tiled=True`. Must be less than or equal to `tile_size`.
+* `sliding_window_size`: Sliding window size for DiT part. Experimental feature, effects are unstable.
+* `sliding_window_stride`: Sliding window stride for DiT part. Experimental feature, effects are unstable.
+* `tea_cache_l1_thresh`: Threshold for TeaCache. Larger values result in faster speed but lower quality. Note that after enabling TeaCache, the inference speed is not uniform, so the remaining time shown on the progress bar becomes inaccurate.
+* `tea_cache_model_id`: TeaCache parameter template, options include `"Wan2.1-T2V-1.3B"`, `"Wan2.1-T2V-14B"`, `"Wan2.1-I2V-14B-480P"`, `"Wan2.1-I2V-14B-720P"`.
+* `progress_bar_cmd`: Progress bar implementation, default is `tqdm.tqdm`. You can set it to `lambda x:x` to disable the progress bar.
+
+</details>
+
+## Model Training
+
+Wan series models are trained using a unified script located at [`./model_training/train.py`](./model_training/train.py).
+
+<details>
+
+<summary>Script Parameters</summary>
+
+The script includes the following parameters:
+
+* Dataset
+  * `--dataset_base_path`: Base path of the dataset.
+  * `--dataset_metadata_path`: Path to the metadata file of the dataset.
+  * `--height`: Height of images or videos. Leave `height` and `width` empty to enable dynamic resolution.
+  * `--width`: Width of images or videos. Leave `height` and `width` empty to enable dynamic resolution.
+  * `--num_frames`: Number of frames per video. Frames are sampled from the video prefix.
+  * `--data_file_keys`: Data file keys in the metadata. Comma-separated.
+  * `--dataset_repeat`: Number of times to repeat the dataset per epoch.
+* Models
+  * `--model_paths`: Paths to load models. In JSON format.
+  * `--model_id_with_origin_paths`: Model ID with origin paths, e.g., Wan-AI/Wan2.1-T2V-1.3B:diffusion_pytorch_model*.safetensors. Comma-separated.
 * Training
   * `--learning_rate`: Learning rate.
   * `--num_epochs`: Number of epochs.
-  * `--output_path`: Save path.
+  * `--output_path`: Output save path.
   * `--remove_prefix_in_ckpt`: Remove prefix in ckpt.
-* Trainable module
+* Trainable Modules
-  * `--trainable_models`: Trainable models, e.g., dit, vae, text_encoder.
+  * `--trainable_models`: Models to train, e.g., dit, vae, text_encoder.
-  * `--lora_base_model`: Add LoRA on which model.
+  * `--lora_base_model`: Which model LoRA is added to.
-  * `--lora_target_modules`: Add LoRA on which layer.
+  * `--lora_target_modules`: Which layers LoRA is added to.
-  * `--lora_rank`: LoRA rank.
+  * `--lora_rank`: Rank of LoRA.
-* Extra model input
+* Extra Inputs
-  * `--input_contains_input_image`: Model input contains `input_image`
+  * `--extra_inputs`: Additional model inputs, comma-separated.
-  * `--input_contains_end_image`: Model input contains `end_image`.
+* VRAM Management
-  * `--input_contains_control_video`: Model input contains `control_video`.
+  * `--use_gradient_checkpointing_offload`: Whether to offload gradient checkpointing to CPU memory.
-  * `--input_contains_reference_image`: Model input contains `reference_image`.
-  * `--input_contains_vace_video`: Model input contains `vace_video`.
-  * `--input_contains_vace_reference_image`: Model input contains `vace_reference_image`.
-  * `--input_contains_motion_bucket_id`: Model input contains `motion_bucket_id`.
 
+Additionally, the training framework is built upon [`accelerate`](https://huggingface.co/docs/accelerate/index). Before starting training, run `accelerate config` to configure GPU-related parameters. For certain training scripts (e.g., full fine-tuning of 14B models), we provide recommended `accelerate` configuration files, which can be found in the corresponding training scripts.
+
+</details>
+
+
+<details>
+
+<summary>Step 1: Prepare the Dataset</summary>
+
+The dataset consists of a series of files. We recommend organizing your dataset as follows:
+
+```
+data/example_video_dataset/
+├── metadata.csv
+├── video1.mp4
+└── video2.mp4
+```
+
+Here, `video1.mp4` and `video2.mp4` are training video files, and `metadata.csv` is the metadata list, for example:
+
+```
+video,prompt
+video1.mp4,"from sunset to night, a small town, light, house, river"
+video2.mp4,"a dog is running"
+```
+
+We have prepared a sample video dataset to help you test. You can download it using the following command:
+
+```shell
+modelscope download --dataset DiffSynth-Studio/example_video_dataset --local_dir ./data/example_video_dataset
+```
+
+The dataset supports mixed training of videos and images. Supported video formats include `"mp4", "avi", "mov", "wmv", "mkv", "flv", "webm"`, and supported image formats include `"jpg", "jpeg", "png", "webp"`.
+
+The resolution of videos can be controlled via script parameters `--height`, `--width`, and `--num_frames`. For each video, the first `num_frames` frames will be used for training; therefore, an error will occur if the video length is less than `num_frames`. Image files will be treated as single-frame videos. When both `--height` and `--width` are left empty, dynamic resolution will be enabled, meaning training will use the actual resolution of each video or image in the dataset.
+
+**We strongly recommend using fixed-resolution training and avoiding mixing images and videos in the same dataset due to load balancing issues in multi-GPU training.**
+
+When the model requires additional inputs, such as the `control_video` needed by control-capable models like [`PAI/Wan2.1-Fun-1.3B-Control`](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control), please add corresponding columns in the metadata file, for example:
+
+```
+video,prompt,control_video
+video1.mp4,"from sunset to night, a small town, light, house, river",video1_softedge.mp4
+```
+
+If additional inputs contain video or image files, their column names need to be specified in the `--data_file_keys` parameter. The default value of this parameter is `"image,video"`, meaning it parses columns named `image` and `video`. You can extend this list based on the additional input requirements, for example: `--data_file_keys "image,video,control_video"`, and also enable `--input_contains_control_video`.
+
+</details>
+
+
+<details>
+
+<summary>Step 2: Load the Model</summary>
+
+Similar to the model loading logic during inference, you can configure the model to be loaded directly via its model ID. For instance, during inference we load the model using:
+
+```python
+model_configs=[
+    ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="diffusion_pytorch_model*.safetensors"),
+    ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth"),
+    ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth"),
+]
+```
+
+During training, simply use the following parameter to load the corresponding model:
+
+```shell
+--model_id_with_origin_paths "Wan-AI/Wan2.1-T2V-1.3B:diffusion_pytorch_model*.safetensors,Wan-AI/Wan2.1-T2V-1.3B:models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.1-T2V-1.3B:Wan2.1_VAE.pth"
+```
+
+If you want to load the model from local files, for example during inference:
+
+```python
+model_configs=[
+    ModelConfig(path=[
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00001-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00002-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00003-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00004-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00005-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00006-of-00006.safetensors",
+    ]),
+    ModelConfig(path="models/Wan-AI/Wan2.1-T2V-14B/models_t5_umt5-xxl-enc-bf16.pth"),
+    ModelConfig(path="models/Wan-AI/Wan2.1-T2V-14B/Wan2.1_VAE.pth"),
+]
+```
+
+Then during training, set the parameter as:
+
+```shell
+--model_paths '[
+    [
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00001-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00002-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00003-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00004-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00005-of-00006.safetensors",
+        "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00006-of-00006.safetensors"
+    ],
+    "models/Wan-AI/Wan2.1-T2V-14B/models_t5_umt5-xxl-enc-bf16.pth",
+    "models/Wan-AI/Wan2.1-T2V-14B/Wan2.1_VAE.pth"
+]' \
+```
+
+</details>
+
+
+<details>
+
+<summary>Step 3: Configure Trainable Modules</summary>
+
+The training framework supports full fine-tuning of base models or LoRA-based training. Here are some examples:
+
+* Full fine-tuning of the DiT module: `--trainable_models dit`
+* Training a LoRA model for the DiT module: `--lora_base_model dit --lora_target_modules "q,k,v,o,ffn.0,ffn.2" --lora_rank 32`
+* Training both a LoRA model for DiT and the Motion Controller (yes, you can train such advanced structures): `--trainable_models motion_controller --lora_base_model dit --lora_target_modules "q,k,v,o,ffn.0,ffn.2" --lora_rank 32`
+
+Additionally, since multiple modules (text encoder, dit, vae) are loaded in the training script, you need to remove prefixes when saving model files. For example, when fully fine-tuning the DiT module or training a LoRA version of DiT, please set `--remove_prefix_in_ckpt pipe.dit.`
+
+</details>
+
+
+<details>
+
+<summary>Step 4: Launch the Training Process</summary>
+
+We have prepared training commands for each model. Please refer to the table at the beginning of this document.
+
+Note that full fine-tuning of the 14B model requires 8 GPUs, each with at least 80GB VRAM. During full fine-tuning of these 14B models, you must install `deepspeed` (`pip install deepspeed`). We have provided recommended [configuration files](./model_training/full/accelerate_config_14B.yaml), which will be loaded automatically in the corresponding training scripts. These scripts have been tested on 8*A100.
+
+The default video resolution in the training script is `480*832*81`. Increasing the resolution may cause out-of-memory errors. To reduce VRAM usage, add the parameter `--use_gradient_checkpointing_offload`.
+
+</details>
+
+## Gallery
+
+1.3B text-to-video:
+
+https://github.com/user-attachments/assets/124397be-cd6a-4f29-a87c-e4c695aaabb8
+
+Put sunglasses on the dog (1.3B video-to-video):
+
+https://github.com/user-attachments/assets/272808d7-fbeb-4747-a6df-14a0860c75fb
+
+14B text-to-video:
+
+https://github.com/user-attachments/assets/3908bc64-d451-485a-8b61-28f6d32dd92f
+
+14B image-to-video:
+
+https://github.com/user-attachments/assets/c0bdd5ca-292f-45ed-b9bc-afe193156e75
+
+LoRA training:
+
+https://github.com/user-attachments/assets/9bd8e30b-97e8-44f9-bb6f-da004ba376a9

examples/wanvideo/README_zh.md

@@ -1,26 +1,44 @@
 # 通义万相 2.1（Wan 2.1）
 
-|模型 ID|类型|额外参数|推理|全量训练|全量训练后验证|LoRA 训练|LoRA 训练后验证|
+[Switch to English](./README.md)
-|-|-|-|-|-|-|-|-|
+
-|[Wan-AI/Wan2.1-T2V-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-1.3B)|基础模型||[code](./model_inference/Wan2.1-T2V-1.3B.py)|[code](./model_training/full/Wan2.1-T2V-1.3B.sh)|[code](./model_training/validate_full/Wan2.1-T2V-1.3B.py)|[code](./model_training/lora/Wan2.1-T2V-1.3B.sh)|[code](./model_training/validate_lora/Wan2.1-T2V-1.3B.py)|
+Wan 2.1 是由阿里巴巴通义实验室开源的一系列视频生成模型。
-|[Wan-AI/Wan2.1-T2V-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-14B)|基础模型||[code](./model_inference/Wan2.1-T2V-14B.py)|[code](./model_training/full/Wan2.1-T2V-14B.sh)|[code](./model_training/validate_full/Wan2.1-T2V-14B.py)|[code](./model_training/lora/Wan2.1-T2V-14B.sh)|[code](./model_training/validate_lora/Wan2.1-T2V-14B.py)|
+
-|[Wan-AI/Wan2.1-I2V-14B-480P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P)|基础模型|`input_image`|[code](./model_inference/Wan2.1-I2V-14B-480P.py)|[code](./model_training/full/Wan2.1-I2V-14B-480P.sh)|[code](./model_training/validate_full/Wan2.1-I2V-14B-480P.py)|[code](./model_training/lora/Wan2.1-I2V-14B-480P.sh)|[code](./model_training/validate_lora/Wan2.1-I2V-14B-480P.py)|
+**DiffSynth-Studio 启用了新的推理和训练框架，如需使用旧版本，请点击[这里](https://github.com/modelscope/DiffSynth-Studio/tree/3edf3583b1f08944cee837b94d9f84d669c2729c)。**
-|[Wan-AI/Wan2.1-I2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-720P)|基础模型|`input_image`|[code](./model_inference/Wan2.1-I2V-14B-720P.py)|[code](./model_training/full/Wan2.1-I2V-14B-720P.sh)|[code](./model_training/validate_full/Wan2.1-I2V-14B-720P.py)|[code](./model_training/lora/Wan2.1-I2V-14B-720P.sh)|[code](./model_training/validate_lora/Wan2.1-I2V-14B-720P.py)|
+
-|[Wan-AI/Wan2.1-FLF2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-FLF2V-14B-720P)|基础模型|`input_image`, `end_image`|[code](./model_inference/Wan2.1-FLF2V-14B-720P.py)|[code](./model_training/full/Wan2.1-FLF2V-14B-720P.sh)|[code](./model_training/validate_full/Wan2.1-FLF2V-14B-720P.py)|[code](./model_training/lora/Wan2.1-FLF2V-14B-720P.sh)|[code](./model_training/validate_lora/Wan2.1-FLF2V-14B-720P.py)|
+## 安装
-|[PAI/Wan2.1-Fun-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP)|基础模型|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-1.3B-InP.py)|[code](./model_training/full/Wan2.1-Fun-1.3B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-1.3B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-1.3B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-1.3B-InP.py)|
+
-|[PAI/Wan2.1-Fun-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)|基础模型|`control_video`|[code](./model_inference/Wan2.1-Fun-1.3B-Control.py)|[code](./model_training/full/Wan2.1-Fun-1.3B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-1.3B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-1.3B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-1.3B-Control.py)|
+在使用本系列模型之前，请通过源码安装 DiffSynth-Studio。
-|[PAI/Wan2.1-Fun-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-InP)|基础模型|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-14B-InP.py)|[code](./model_training/full/Wan2.1-Fun-14B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-14B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-14B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-14B-InP.py)|
+
-|[PAI/Wan2.1-Fun-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-Control)|基础模型|`control_video`|[code](./model_inference/Wan2.1-Fun-14B-Control.py)|[code](./model_training/full/Wan2.1-Fun-14B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-14B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-14B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-14B-Control.py)|
+```shell
-|[PAI/Wan2.1-Fun-V1.1-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)|基础模型|`control_video`, `reference_image`|[code](./model_inference/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control.py)|
+git clone https://github.com/modelscope/DiffSynth-Studio.git
-|[PAI/Wan2.1-Fun-V1.1-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control)|基础模型|`control_video`, `reference_image`|[code](./model_inference/Wan2.1-Fun-V1.1-14B-Control.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control.py)|
+cd DiffSynth-Studio
-|[PAI/Wan2.1-Fun-V1.1-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-InP)|基础模型|`input_image`, `end_image`||||||
+pip install -e .
-|[PAI/Wan2.1-Fun-V1.1-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP)|基础模型|`input_image`, `end_image`||||||
+```
-|[PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera)|基础模型|||||||
+
-|[PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)|基础模型|||||||
+## 模型总览
-|[iic/VACE-Wan2.1-1.3B-Preview](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)|适配器|`vace_control_video`, `vace_reference_image`|[code](./model_inference/Wan2.1-VACE-1.3B-Preview.py)|[code](./model_training/full/VACE-Wan2.1-1.3B-Preview.sh)|[code](./model_training/validate_full/VACE-Wan2.1-1.3B-Preview.py)|[code](./model_training/lora/VACE-Wan2.1-1.3B-Preview.sh)|[code](./model_training/validate_lora/VACE-Wan2.1-1.3B-Preview.py)|
+
-|[Wan-AI/Wan2.1-VACE-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-1.3B)|适配器|`vace_control_video`, `vace_reference_image`||||||
+|模型 ID|额外参数|推理|全量训练|全量训练后验证|LoRA 训练|LoRA 训练后验证|
-|[Wan-AI/Wan2.1-VACE-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-14B)|适配器|`vace_control_video`, `vace_reference_image`||||||
+|-|-|-|-|-|-|-|
-|[DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1)|适配器|`motion_bucket_id`|[code](./model_inference/Wan2.1-1.3b-speedcontrol-v1.py)|[code](./model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](./model_training/validate_full/Wan2.1-1.3b-speedcontrol-v1.py)|[code](./model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](./model_training/validate_lora/Wan2.1-1.3b-speedcontrol-v1.py)|
+|[Wan-AI/Wan2.1-T2V-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-1.3B)||[code](./model_inference/Wan2.1-T2V-1.3B.py)|[code](./model_training/full/Wan2.1-T2V-1.3B.sh)|[code](./model_training/validate_full/Wan2.1-T2V-1.3B.py)|[code](./model_training/lora/Wan2.1-T2V-1.3B.sh)|[code](./model_training/validate_lora/Wan2.1-T2V-1.3B.py)|
+|[Wan-AI/Wan2.1-T2V-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-14B)||[code](./model_inference/Wan2.1-T2V-14B.py)|[code](./model_training/full/Wan2.1-T2V-14B.sh)|[code](./model_training/validate_full/Wan2.1-T2V-14B.py)|[code](./model_training/lora/Wan2.1-T2V-14B.sh)|[code](./model_training/validate_lora/Wan2.1-T2V-14B.py)|
+|[Wan-AI/Wan2.1-I2V-14B-480P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P)|`input_image`|[code](./model_inference/Wan2.1-I2V-14B-480P.py)|[code](./model_training/full/Wan2.1-I2V-14B-480P.sh)|[code](./model_training/validate_full/Wan2.1-I2V-14B-480P.py)|[code](./model_training/lora/Wan2.1-I2V-14B-480P.sh)|[code](./model_training/validate_lora/Wan2.1-I2V-14B-480P.py)|
+|[Wan-AI/Wan2.1-I2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-720P)|`input_image`|[code](./model_inference/Wan2.1-I2V-14B-720P.py)|[code](./model_training/full/Wan2.1-I2V-14B-720P.sh)|[code](./model_training/validate_full/Wan2.1-I2V-14B-720P.py)|[code](./model_training/lora/Wan2.1-I2V-14B-720P.sh)|[code](./model_training/validate_lora/Wan2.1-I2V-14B-720P.py)|
+|[Wan-AI/Wan2.1-FLF2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-FLF2V-14B-720P)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-FLF2V-14B-720P.py)|[code](./model_training/full/Wan2.1-FLF2V-14B-720P.sh)|[code](./model_training/validate_full/Wan2.1-FLF2V-14B-720P.py)|[code](./model_training/lora/Wan2.1-FLF2V-14B-720P.sh)|[code](./model_training/validate_lora/Wan2.1-FLF2V-14B-720P.py)|
+|[PAI/Wan2.1-Fun-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-1.3B-InP.py)|[code](./model_training/full/Wan2.1-Fun-1.3B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-1.3B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-1.3B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)|`control_video`|[code](./model_inference/Wan2.1-Fun-1.3B-Control.py)|[code](./model_training/full/Wan2.1-Fun-1.3B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-1.3B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-1.3B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-14B-InP.py)|[code](./model_training/full/Wan2.1-Fun-14B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-14B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-14B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-14B-InP.py)|
+|[PAI/Wan2.1-Fun-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-Control)|`control_video`|[code](./model_inference/Wan2.1-Fun-14B-Control.py)|[code](./model_training/full/Wan2.1-Fun-14B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-14B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-14B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)|`control_video`, `reference_image`|[code](./model_inference/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control)|`control_video`, `reference_image`|[code](./model_inference/Wan2.1-Fun-V1.1-14B-Control.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP)|`input_image`, `end_image`|[code](./model_inference/Wan2.1-Fun-V1.1-14B-InP.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-14B-InP.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-14B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera)|`control_camera_video`, `input_image`|[code](./model_inference/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)|`control_camera_video`, `input_image`|[code](./model_inference/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](./model_training/full/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](./model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](./model_training/lora/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](./model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|
+|[iic/VACE-Wan2.1-1.3B-Preview](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)|`vace_control_video`, `vace_reference_image`|[code](./model_inference/Wan2.1-VACE-1.3B-Preview.py)|[code](./model_training/full/Wan2.1-VACE-1.3B-Preview.sh)|[code](./model_training/validate_full/Wan2.1-VACE-1.3B-Preview.py)|[code](./model_training/lora/Wan2.1-VACE-1.3B-Preview.sh)|[code](./model_training/validate_lora/Wan2.1-VACE-1.3B-Preview.py)|
+|[Wan-AI/Wan2.1-VACE-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-1.3B)|`vace_control_video`, `vace_reference_image`|[code](./model_inference/Wan2.1-VACE-1.3B.py)|[code](./model_training/full/Wan2.1-VACE-1.3B.sh)|[code](./model_training/validate_full/Wan2.1-VACE-1.3B.py)|[code](./model_training/lora/Wan2.1-VACE-1.3B.sh)|[code](./model_training/validate_lora/Wan2.1-VACE-1.3B.py)|
+|[Wan-AI/Wan2.1-VACE-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-14B)|`vace_control_video`, `vace_reference_image`|[code](./model_inference/Wan2.1-VACE-14B.py)|[code](./model_training/full/Wan2.1-VACE-14B.sh)|[code](./model_training/validate_full/Wan2.1-VACE-14B.py)|[code](./model_training/lora/Wan2.1-VACE-14B.sh)|[code](./model_training/validate_lora/Wan2.1-VACE-14B.py)|
+|[DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1)|`motion_bucket_id`|[code](./model_inference/Wan2.1-1.3b-speedcontrol-v1.py)|[code](./model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](./model_training/validate_full/Wan2.1-1.3b-speedcontrol-v1.py)|[code](./model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](./model_training/validate_lora/Wan2.1-1.3b-speedcontrol-v1.py)|
 
 ## 模型推理
 
@@ -78,6 +96,7 @@ ModelConfig(path=[
 * `local_model_path`: 用于保存下载模型的路径，默认值为 `"./models"`。
 * `skip_download`: 是否跳过下载，默认值为 `False`。当您的网络无法访问[魔搭社区](https://modelscope.cn/)时，请手动下载必要的文件，并将其设置为 `True`。
 * `redirect_common_files`: 是否重定向重复模型文件，默认值为 `True`。由于 Wan 系列模型包括多个基础模型，每个基础模型的 text encoder 等模块都是相同的，为避免重复下载，我们会对模型路径进行重定向。
+* `use_usp`: 是否启用 Unified Sequence Parallel，默认值为 `False`。用于多 GPU 并行推理。
 
 </details>
 
@@ -142,6 +161,29 @@ FP8 量化能够大幅度减少显存占用，但不会加速，部分模型在
 </details>
 
 
+<details>
+
+<summary>推理加速</summary>
+
+Wan 支持多种加速方案，包括
+
+* 高效注意力机制实现：当您的 Python 环境中安装过这些注意力机制实现方案时，我们将会按照以下优先级自动启用。
+    * [Flash Attention 3](https://github.com/Dao-AILab/flash-attention)
+    * [Flash Attention 2](https://github.com/Dao-AILab/flash-attention)
+    * [Sage Attention](https://github.com/thu-ml/SageAttention)
+    * [torch SDPA](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html) (默认设置，建议安装 `torch>=2.5.0`)
+* 统一序列并行：基于 [xDiT](https://github.com/xdit-project/xDiT) 实现的序列并行，请参考[示例代码](./acceleration/unified_sequence_parallel.py)，使用以下命令运行：
+
+```shell
+pip install xfuser>=0.4.3
+torchrun --standalone --nproc_per_node=8 examples/wanvideo/acceleration/unified_sequence_parallel.py
+```
+
+* TeaCache：加速技术 [TeaCache](https://github.com/ali-vilab/TeaCache)，请参考[示例代码](./acceleration/teacache.py)。
+
+</details>
+
+
 <details>
 
 <summary>输入参数</summary>
@@ -156,6 +198,9 @@ Pipeline 在推理阶段能够接收以下输入参数：
 * `denoising_strength`: 去噪强度，范围为 [0, 1]。数值越小，生成的视频越接近 `input_video`。
 * `control_video`: 控制视频，适用于带控制能力的 Wan 系列模型，例如 [`PAI/Wan2.1-Fun-1.3B-Control`](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)。
 * `reference_image`: 参考图片，适用于带参考图能力的 Wan 系列模型，例如 [`PAI/Wan2.1-Fun-V1.1-1.3B-Control`](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)。
+* `camera_control_direction`: 镜头控制方向，可选 "Left", "Right", "Up", "Down", "LeftUp", "LeftDown", "RightUp", "RightDown" 之一，适用于 Camera-Control 模型，例如 [PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://www.modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)。
+* `camera_control_speed`: 镜头控制速度，适用于 Camera-Control 模型，例如 [PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://www.modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)。
+* `camera_control_origin`: 镜头控制序列的原点坐标，请参考[原论文](https://arxiv.org/pdf/2404.02101)进行设置，适用于 Camera-Control 模型，例如 [PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://www.modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)。
 * `vace_video`: VACE 模型的输入视频，适用于 VACE 系列模型，例如 [`iic/VACE-Wan2.1-1.3B-Preview`](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)。
 * `vace_video_mask`: VACE 模型的 mask 视频，适用于 VACE 系列模型，例如 [`iic/VACE-Wan2.1-1.3B-Preview`](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)。
 * `vace_reference_image`: VACE 模型的参考图片，适用于 VACE 系列模型，例如 [`iic/VACE-Wan2.1-1.3B-Preview`](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)。
@@ -214,16 +259,12 @@ Wan 系列模型训练通过统一的 [`./model_training/train.py`](./model_trai
   * `--lora_target_modules`: LoRA 添加到哪一层上。
   * `--lora_rank`: LoRA 的秩（Rank）。
 * 额外模型输入
-  * `--input_contains_input_image`: 模型输入包含 `input_image`
+  * `--extra_inputs`: 额外的模型输入，以逗号分隔。
-  * `--input_contains_end_image`: 模型输入包含 `end_image`。
-  * `--input_contains_control_video`: 模型输入包含 `control_video`。
-  * `--input_contains_reference_image`: 模型输入包含 `reference_image`。
-  * `--input_contains_vace_video`: 模型输入包含 `vace_video`。
-  * `--input_contains_vace_reference_image`: 模型输入包含 `vace_reference_image`。
-  * `--input_contains_motion_bucket_id`: 模型输入包含 `motion_bucket_id`。
 * 显存管理
   * `--use_gradient_checkpointing_offload`: 是否将 gradient checkpointing 卸载到内存中。
 
+此外，训练框架基于 [`accelerate`](https://huggingface.co/docs/accelerate/index) 构建，在开始训练前运行 `accelerate config` 可配置 GPU 的相关参数。对于部分模型训练（例如 14B 模型的全量训练）脚本，我们提供了建议的 `accelerate` 配置文件，可在对应的训练脚本中查看。
+
 </details>
 
 
@@ -251,7 +292,7 @@ video2.mp4,"a dog is running"
 我们构建了一个样例视频数据集，以方便您进行测试，通过以下命令可以下载这个数据集：
 
 ```shell
-modelscope download --dataset DiffSynth-Studio/example_video_dataset README.md --local_dir ./data/example_video_dataset
+modelscope download --dataset DiffSynth-Studio/example_video_dataset --local_dir ./data/example_video_dataset
 ```
 
 数据集支持视频和图片混合训练，支持的视频文件格式包括 `"mp4", "avi", "mov", "wmv", "mkv", "flv", "webm"`，支持的图片格式包括 `"jpg", "jpeg", "png", "webp"`。
@@ -355,3 +396,25 @@ model_configs=[
 训练脚本的默认视频尺寸为 `480*832*81`，提升分辨率将可能导致显存不足，请添加参数 `--use_gradient_checkpointing_offload` 降低显存占用。
 
 </details>
+
+## 案例展示
+
+1.3B 文生视频：
+
+https://github.com/user-attachments/assets/124397be-cd6a-4f29-a87c-e4c695aaabb8
+
+给狗狗戴上墨镜（1.3B 视频生视频）：
+
+https://github.com/user-attachments/assets/272808d7-fbeb-4747-a6df-14a0860c75fb
+
+14B 文生视频：
+
+https://github.com/user-attachments/assets/3908bc64-d451-485a-8b61-28f6d32dd92f
+
+14B 图生视频：
+
+https://github.com/user-attachments/assets/c0bdd5ca-292f-45ed-b9bc-afe193156e75
+
+LoRA 训练：
+
+https://github.com/user-attachments/assets/9bd8e30b-97e8-44f9-bb6f-da004ba376a9

examples/wanvideo/acceleration/teacache.py

@@ -1,34 +1,27 @@
 import torch
-from diffsynth import ModelManager, WanVideoPipeline, save_video, VideoData
+from PIL import Image
-from modelscope import snapshot_download
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
 
 
-# Download models
+pipe = WanVideoPipeline.from_pretrained(
-snapshot_download("Wan-AI/Wan2.1-T2V-1.3B", local_dir="models/Wan-AI/Wan2.1-T2V-1.3B")
+    torch_dtype=torch.bfloat16,
-
+    device="cuda",
-# Load models
+    model_configs=[
-model_manager = ModelManager(device="cpu")
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
-model_manager.load_models(
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
-    [
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
-        "models/Wan-AI/Wan2.1-T2V-1.3B/diffusion_pytorch_model.safetensors",
-        "models/Wan-AI/Wan2.1-T2V-1.3B/models_t5_umt5-xxl-enc-bf16.pth",
-        "models/Wan-AI/Wan2.1-T2V-1.3B/Wan2.1_VAE.pth",
     ],
-    torch_dtype=torch.bfloat16, # You can set `torch_dtype=torch.float8_e4m3fn` to enable FP8 quantization.
 )
-pipe = WanVideoPipeline.from_model_manager(model_manager, torch_dtype=torch.bfloat16, device="cuda")
+pipe.enable_vram_management()
-pipe.enable_vram_management(num_persistent_param_in_dit=None)
+
 
-# Text-to-video
 video = pipe(
     prompt="纪实摄影风格画面，一只活泼的小狗在绿茵茵的草地上迅速奔跑。小狗毛色棕黄，两只耳朵立起，神情专注而欢快。阳光洒在它身上，使得毛发看上去格外柔软而闪亮。背景是一片开阔的草地，偶尔点缀着几朵野花，远处隐约可见蓝天和几片白云。透视感鲜明，捕捉小狗奔跑时的动感和四周草地的生机。中景侧面移动视角。",
     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",
-    num_inference_steps=50,
     seed=0, tiled=True,
     # TeaCache parameters
     tea_cache_l1_thresh=0.05, # The larger this value is, the faster the speed, but the worse the visual quality.
     tea_cache_model_id="Wan2.1-T2V-1.3B", # Choose one in (Wan2.1-T2V-1.3B, Wan2.1-T2V-14B, Wan2.1-I2V-14B-480P, Wan2.1-I2V-14B-720P).
 )
 save_video(video, "video1.mp4", fps=15, quality=5)
-
-# TeaCache doesn't support video-to-video

examples/wanvideo/acceleration/unified_sequence_parallel.py

@@ -0,0 +1,27 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+import torch.distributed as dist
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    use_usp=True,
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+pipe.enable_vram_management()
+
+
+video = pipe(
+    prompt="一名宇航员身穿太空服，面朝镜头骑着一匹机械马在火星表面驰骋。红色的荒凉地表延伸至远方，点缀着巨大的陨石坑和奇特的岩石结构。机械马的步伐稳健，扬起微弱的尘埃，展现出未来科技与原始探索的完美结合。宇航员手持操控装置，目光坚定，仿佛正在开辟人类的新疆域。背景是深邃的宇宙和蔚蓝的地球，画面既科幻又充满希望，让人不禁畅想未来的星际生活。",
+    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,
+)
+if dist.get_rank() == 0:
+    save_video(video, "video1.mp4", fps=15, quality=5)

examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py

@@ -0,0 +1,44 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.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)

examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-InP.py

@@ -0,0 +1,36 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.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)

examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control-Camera.py

@@ -0,0 +1,44 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.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)

examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-InP.py

@@ -0,0 +1,36 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.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)

examples/wanvideo/model_inference/Wan2.1-VACE-1.3B.py

@@ -0,0 +1,53 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", 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/depth_video.mp4", "data/examples/wan/cat_fightning.jpg"]
+)
+
+# Depth video -> Video
+control_video = VideoData("data/examples/wan/depth_video.mp4", height=480, width=832)
+video = pipe(
+    prompt="两只可爱的橘猫戴上拳击手套，站在一个拳击台上搏斗。",
+    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",
+    vace_video=control_video,
+    seed=1, tiled=True
+)
+save_video(video, "video1.mp4", fps=15, quality=5)
+
+# Reference image -> Video
+video = pipe(
+    prompt="两只可爱的橘猫戴上拳击手套，站在一个拳击台上搏斗。",
+    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",
+    vace_reference_image=Image.open("data/examples/wan/cat_fightning.jpg").resize((832, 480)),
+    seed=1, tiled=True
+)
+save_video(video, "video2.mp4", fps=15, quality=5)
+
+# Depth video + Reference image -> Video
+video = pipe(
+    prompt="两只可爱的橘猫戴上拳击手套，站在一个拳击台上搏斗。",
+    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",
+    vace_video=control_video,
+    vace_reference_image=Image.open("data/examples/wan/cat_fightning.jpg").resize((832, 480)),
+    seed=1, tiled=True
+)
+save_video(video, "video3.mp4", fps=15, quality=5)

examples/wanvideo/model_inference/Wan2.1-VACE-14B.py

@@ -0,0 +1,54 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", 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/depth_video.mp4", "data/examples/wan/cat_fightning.jpg"]
+)
+
+# Depth video -> Video
+control_video = VideoData("data/examples/wan/depth_video.mp4", height=480, width=832)
+video = pipe(
+    prompt="两只可爱的橘猫戴上拳击手套，站在一个拳击台上搏斗。",
+    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",
+    vace_video=control_video,
+    seed=1, tiled=True
+)
+save_video(video, "video1_14b.mp4", fps=15, quality=5)
+
+# Reference image -> Video
+video = pipe(
+    prompt="两只可爱的橘猫戴上拳击手套，站在一个拳击台上搏斗。",
+    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",
+    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)
+
+# Depth video + Reference image -> Video
+video = pipe(
+    prompt="两只可爱的橘猫戴上拳击手套，站在一个拳击台上搏斗。",
+    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",
+    vace_video=control_video,
+    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)

examples/wanvideo/model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh

@@ -10,4 +10,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --remove_prefix_in_ckpt "pipe.motion_controller." \
   --output_path "./models/train/Wan2.1-1.3b-speedcontrol-v1_full" \
   --trainable_models "motion_controller" \
-  --input_contains_motion_bucket_id
+  --extra_inputs "motion_bucket_id"

examples/wanvideo/model_training/full/Wan2.1-FLF2V-14B-720P.sh

@@ -10,5 +10,4 @@ accelerate launch --config_file examples/wanvideo/model_training/full/accelerate
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-FLF2V-14B-720P_full" \
   --trainable_models "dit" \
-  --input_contains_input_image \
+  --extra_inputs "input_image,end_image"
-  --input_contains_end_image

examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-Control.sh

@@ -11,4 +11,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-Fun-1.3B-Control_full" \
   --trainable_models "dit" \
-  --input_contains_control_video
+  --extra_inputs "control_video"

examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-InP.sh

@@ -10,5 +10,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-Fun-1.3B-InP_full" \
   --trainable_models "dit" \
-  --input_contains_input_image \
+  --extra_inputs "input_image,end_image"
-  --input_contains_end_image

examples/wanvideo/model_training/full/Wan2.1-Fun-14B-Control.sh

@@ -11,4 +11,4 @@ accelerate launch --config_file examples/wanvideo/model_training/full/accelerate
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-Fun-14B-Control_full" \
   --trainable_models "dit" \
-  --input_contains_control_video
+  --extra_inputs "control_video"

examples/wanvideo/model_training/full/Wan2.1-Fun-14B-InP.sh

@@ -10,5 +10,4 @@ accelerate launch --config_file examples/wanvideo/model_training/full/accelerate
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-Fun-14B-InP_full" \
   --trainable_models "dit" \
-  --input_contains_input_image \
+  --extra_inputs "input_image,end_image"
-  --input_contains_end_image

examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh

@@ -0,0 +1,13 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_camera_control.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-1.3B-Control-Camera_full" \
+  --trainable_models "dit" \
+  --extra_inputs "input_image,camera_control_direction,camera_control_speed"

examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh

@@ -11,5 +11,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-Fun-V1.1-1.3B-Control_full" \
   --trainable_models "dit" \
-  --input_contains_control_video \
+  --extra_inputs "control_video,reference_image"
-  --input_contains_reference_image

examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-InP.sh

@@ -0,0 +1,13 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-1.3B-InP:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-1.3B-InP:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-1.3B-InP:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-1.3B-InP:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-1.3B-InP_full" \
+  --trainable_models "dit" \
+  --extra_inputs "input_image,end_image"

examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control-Camera.sh

@@ -0,0 +1,13 @@
+accelerate launch --config_file examples/wanvideo/model_training/full/accelerate_config_14B.yaml examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_camera_control.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-14B-Control-Camera_full" \
+  --trainable_models "dit" \
+  --extra_inputs "input_image,camera_control_direction,camera_control_speed"

examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh

@@ -11,5 +11,4 @@ accelerate launch --config_file examples/wanvideo/model_training/full/accelerate
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-Fun-V1.1-14B-Control_full" \
   --trainable_models "dit" \
-  --input_contains_control_video \
+  --extra_inputs "control_video,reference_image"
-  --input_contains_reference_image

examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-InP.sh

@@ -0,0 +1,13 @@
+accelerate launch --config_file examples/wanvideo/model_training/full/accelerate_config_14B.yaml examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-14B-InP:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-14B-InP:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-14B-InP:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-14B-InP:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-14B-InP_full" \
+  --trainable_models "dit" \
+  --extra_inputs "input_image,end_image"

examples/wanvideo/model_training/full/Wan2.1-I2V-14B-480P.sh

@@ -10,4 +10,4 @@ accelerate launch --config_file examples/wanvideo/model_training/full/accelerate
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-I2V-14B-480P_full" \
   --trainable_models "dit" \
-  --input_contains_input_image
+  --extra_inputs "input_image"

examples/wanvideo/model_training/full/Wan2.1-I2V-14B-720P.sh

@@ -10,4 +10,4 @@ accelerate launch --config_file examples/wanvideo/model_training/full/accelerate
   --remove_prefix_in_ckpt "pipe.dit." \
   --output_path "./models/train/Wan2.1-I2V-14B-720P_full" \
   --trainable_models "dit" \
-  --input_contains_input_image
+  --extra_inputs "input_image"

examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B-Preview.sh

@@ -0,0 +1,16 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_vace.csv \
+  --data_file_keys "video,vace_video,vace_reference_image" \
+  --height 480 \
+  --width 832 \
+  --num_frames 49 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "iic/VACE-Wan2.1-1.3B-Preview:diffusion_pytorch_model*.safetensors,iic/VACE-Wan2.1-1.3B-Preview:models_t5_umt5-xxl-enc-bf16.pth,iic/VACE-Wan2.1-1.3B-Preview:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.vace." \
+  --output_path "./models/train/Wan2.1-VACE-1.3B-Preview_full" \
+  --trainable_models "vace" \
+  --extra_inputs "vace_video,vace_reference_image" \
+  --use_gradient_checkpointing_offload

examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B.sh

@@ -0,0 +1,16 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_vace.csv \
+  --data_file_keys "video,vace_video,vace_reference_image" \
+  --height 480 \
+  --width 832 \
+  --num_frames 49 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.1-VACE-1.3B:diffusion_pytorch_model*.safetensors,Wan-AI/Wan2.1-VACE-1.3B:models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.1-VACE-1.3B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.vace." \
+  --output_path "./models/train/Wan2.1-VACE-1.3B_full" \
+  --trainable_models "vace" \
+  --extra_inputs "vace_video,vace_reference_image" \
+  --use_gradient_checkpointing_offload

examples/wanvideo/model_training/full/Wan2.1-VACE-14B.sh

@@ -0,0 +1,16 @@
+accelerate launch --config_file examples/wanvideo/model_training/full/accelerate_config_14B.yaml examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_vace.csv \
+  --data_file_keys "video,vace_video,vace_reference_image" \
+  --height 480 \
+  --width 832 \
+  --num_frames 17 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.1-VACE-14B:diffusion_pytorch_model*.safetensors,Wan-AI/Wan2.1-VACE-14B:models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.1-VACE-14B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.vace." \
+  --output_path "./models/train/Wan2.1-VACE-14B_full" \
+  --trainable_models "vace" \
+  --extra_inputs "vace_video,vace_reference_image" \
+  --use_gradient_checkpointing_offload

examples/wanvideo/model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh

@@ -12,4 +12,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_motion_bucket_id
+  --extra_inputs "motion_bucket_id"

examples/wanvideo/model_training/lora/Wan2.1-FLF2V-14B-720P.sh

@@ -12,5 +12,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_input_image \
+  --extra_inputs "input_image,end_image"
-  --input_contains_end_image

examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-Control.sh

@@ -13,4 +13,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_control_video
+  --extra_inputs "control_video"

examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-InP.sh

@@ -12,5 +12,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_input_image \
+  --extra_inputs "input_image,end_image"
-  --input_contains_end_image

examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-Control.sh

@@ -13,4 +13,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_control_video
+  --extra_inputs "control_video"

examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-InP.sh

@@ -12,5 +12,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_input_image \
+  --extra_inputs "input_image,end_image"
-  --input_contains_end_image

examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh

@@ -0,0 +1,15 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_camera_control.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-1.3B-Control-Camera_lora" \
+  --lora_base_model "dit" \
+  --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
+  --lora_rank 32 \
+  --extra_inputs "input_image,camera_control_direction,camera_control_speed"

examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh

@@ -13,5 +13,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_control_video \
+  --extra_inputs "control_video,reference_image"
-  --input_contains_reference_image

examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-InP.sh

@@ -0,0 +1,15 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-1.3B-InP:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-1.3B-InP:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-1.3B-InP:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-1.3B-InP:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-1.3B-InP_lora" \
+  --lora_base_model "dit" \
+  --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
+  --lora_rank 32 \
+  --extra_inputs "input_image,end_image"

examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control-Camera.sh

@@ -0,0 +1,15 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_camera_control.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-14B-Control-Camera:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-14B-Control-Camera_lora" \
+  --lora_base_model "dit" \
+  --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
+  --lora_rank 32 \
+  --extra_inputs "input_image,camera_control_direction,camera_control_speed"

examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh

@@ -13,5 +13,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_control_video \
+  --extra_inputs "control_video,reference_image"
-  --input_contains_reference_image

examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-InP.sh

@@ -0,0 +1,15 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata.csv \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "PAI/Wan2.1-Fun-V1.1-14B-InP:diffusion_pytorch_model*.safetensors,PAI/Wan2.1-Fun-V1.1-14B-InP:models_t5_umt5-xxl-enc-bf16.pth,PAI/Wan2.1-Fun-V1.1-14B-InP:Wan2.1_VAE.pth,PAI/Wan2.1-Fun-V1.1-14B-InP:models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.1-Fun-V1.1-14B-InP_lora" \
+  --lora_base_model "dit" \
+  --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
+  --lora_rank 32 \
+  --extra_inputs "input_image,end_image"

examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-480P.sh

@@ -12,4 +12,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_input_image
+  --extra_inputs "input_image"

examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-720P.sh

@@ -12,4 +12,4 @@ accelerate launch examples/wanvideo/model_training/train.py \
   --lora_base_model "dit" \
   --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
   --lora_rank 32 \
-  --input_contains_input_image
+  --extra_inputs "input_image"

examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B-Preview.sh

@@ -0,0 +1,17 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_vace.csv \
+  --data_file_keys "video,vace_video,vace_reference_image" \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "iic/VACE-Wan2.1-1.3B-Preview:diffusion_pytorch_model*.safetensors,iic/VACE-Wan2.1-1.3B-Preview:models_t5_umt5-xxl-enc-bf16.pth,iic/VACE-Wan2.1-1.3B-Preview:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.vace." \
+  --output_path "./models/train/Wan2.1-VACE-1.3B-Preview_lora" \
+  --lora_base_model "vace" \
+  --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
+  --lora_rank 32 \
+  --extra_inputs "vace_video,vace_reference_image" \
+  --use_gradient_checkpointing_offload

examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B.sh

@@ -0,0 +1,17 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_vace.csv \
+  --data_file_keys "video,vace_video,vace_reference_image" \
+  --height 480 \
+  --width 832 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.1-VACE-1.3B:diffusion_pytorch_model*.safetensors,Wan-AI/Wan2.1-VACE-1.3B:models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.1-VACE-1.3B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.vace." \
+  --output_path "./models/train/Wan2.1-VACE-1.3B_lora" \
+  --lora_base_model "vace" \
+  --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
+  --lora_rank 32 \
+  --extra_inputs "vace_video,vace_reference_image" \
+  --use_gradient_checkpointing_offload

examples/wanvideo/model_training/lora/Wan2.1-VACE-14B.sh

@@ -0,0 +1,18 @@
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/example_video_dataset \
+  --dataset_metadata_path data/example_video_dataset/metadata_vace.csv \
+  --data_file_keys "video,vace_video,vace_reference_image" \
+  --height 480 \
+  --width 832 \
+  --num_frames 17 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.1-VACE-14B:diffusion_pytorch_model*.safetensors,Wan-AI/Wan2.1-VACE-14B:models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.1-VACE-14B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.vace." \
+  --output_path "./models/train/Wan2.1-VACE-14B_lora" \
+  --lora_base_model "vace" \
+  --lora_target_modules "q,k,v,o,ffn.0,ffn.2" \
+  --lora_rank 32 \
+  --extra_inputs "vace_video,vace_reference_image" \
+  --use_gradient_checkpointing_offload

examples/wanvideo/model_training/train.py

@@ -1,6 +1,6 @@
 import torch, os, json
 from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
-from diffsynth.trainers.utils import DiffusionTrainingModule, VideoDataset, launch_training_task, wan_parser
+from diffsynth.trainers.utils import DiffusionTrainingModule, VideoDataset, ModelLogger, launch_training_task, wan_parser
 os.environ["TOKENIZERS_PARALLELISM"] = "false"
 
 
@@ -13,14 +13,7 @@ class WanTrainingModule(DiffusionTrainingModule):
         lora_base_model=None, lora_target_modules="q,k,v,o,ffn.0,ffn.2", lora_rank=32,
         use_gradient_checkpointing=True,
         use_gradient_checkpointing_offload=False,
-        # Extra inputs
+        extra_inputs=None,
-        input_contains_input_image=False,
-        input_contains_end_image=False,
-        input_contains_control_video=False,
-        input_contains_reference_image=False,
-        input_contains_vace_video=False,
-        input_contains_vace_reference_image=False,
-        input_contains_motion_bucket_id=False,
     ):
         super().__init__()
         # Load models
@@ -51,13 +44,7 @@ class WanTrainingModule(DiffusionTrainingModule):
         # Store other configs
         self.use_gradient_checkpointing = use_gradient_checkpointing
         self.use_gradient_checkpointing_offload = use_gradient_checkpointing_offload
-        self.input_contains_input_image = input_contains_input_image
+        self.extra_inputs = extra_inputs.split(",") if extra_inputs is not None else []
-        self.input_contains_end_image = input_contains_end_image
-        self.input_contains_control_video = input_contains_control_video
-        self.input_contains_reference_image = input_contains_reference_image
-        self.input_contains_vace_video = input_contains_vace_video
-        self.input_contains_vace_reference_image = input_contains_vace_reference_image
-        self.input_contains_motion_bucket_id = input_contains_motion_bucket_id
         
         
     def forward_preprocess(self, data):
@@ -85,13 +72,13 @@ class WanTrainingModule(DiffusionTrainingModule):
         }
         
         # Extra inputs
-        if self.input_contains_input_image: inputs_shared["input_image"] = data["video"][0]
+        for extra_input in self.extra_inputs:
-        if self.input_contains_end_image: inputs_shared["end_image"] = data["video"][-1]
+            if extra_input == "input_image":
-        if self.input_contains_control_video: inputs_shared["control_video"] = data["control_video"]
+                inputs_shared["input_image"] = data["video"][0]
-        if self.input_contains_reference_image: inputs_shared["reference_image"] = data["reference_image"]
+            elif extra_input == "end_image":
-        if self.input_contains_vace_video: inputs_shared["vace_video"] = data["vace_video"]
+                inputs_shared["end_image"] = data["video"][-1]
-        if self.input_contains_vace_reference_image: inputs_shared["vace_reference_image"] = data["vace_reference_image"]
+            else:
-        if self.input_contains_motion_bucket_id: inputs_shared["motion_bucket_id"] = data["motion_bucket_id"]
+                inputs_shared[extra_input] = data[extra_input]
         
         # Pipeline units will automatically process the input parameters.
         for unit in self.pipe.units:
@@ -118,12 +105,16 @@ if __name__ == "__main__":
         lora_target_modules=args.lora_target_modules,
         lora_rank=args.lora_rank,
         use_gradient_checkpointing_offload=args.use_gradient_checkpointing_offload,
-        input_contains_input_image=args.input_contains_input_image,
+        extra_inputs=args.extra_inputs,
-        input_contains_end_image=args.input_contains_end_image,
+    )
-        input_contains_control_video=args.input_contains_control_video,
+    model_logger = ModelLogger(
-        input_contains_reference_image=args.input_contains_reference_image,
+        args.output_path,
-        input_contains_vace_video=args.input_contains_vace_video,
+        remove_prefix_in_ckpt=args.remove_prefix_in_ckpt
-        input_contains_vace_reference_image=args.input_contains_vace_reference_image,
+    )
-        input_contains_motion_bucket_id=args.input_contains_motion_bucket_id,
+    optimizer = torch.optim.AdamW(model.trainable_modules(), lr=args.learning_rate)
+    scheduler = torch.optim.lr_scheduler.ConstantLR(optimizer)
+    launch_training_task(
+        dataset, model, model_logger, optimizer, scheduler,
+        num_epochs=args.num_epochs,
+        gradient_accumulation_steps=args.gradient_accumulation_steps,
     )
-    launch_training_task(model, dataset, args=args)

examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py

@@ -0,0 +1,32 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+state_dict = load_state_dict("models/train/Wan2.1-Fun-V1.1-1.3B-Control-Camera_full/epoch-1.safetensors")
+pipe.dit.load_state_dict(state_dict)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0],
+    camera_control_direction="Left", camera_control_speed=0.0,
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-1.3B-Control-Camera.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-InP.py

@@ -0,0 +1,31 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+state_dict = load_state_dict("models/train/Wan2.1-Fun-V1.1-1.3B-InP_full/epoch-1.safetensors")
+pipe.dit.load_state_dict(state_dict)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0], end_image=video[80],
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-1.3B-InP.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control-Camera.py

@@ -0,0 +1,32 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+state_dict = load_state_dict("models/train/Wan2.1-Fun-V1.1-14B-Control-Camera_full/epoch-1.safetensors")
+pipe.dit.load_state_dict(state_dict)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0],
+    camera_control_direction="Left", camera_control_speed=0.0,
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-14B-Control-Camera.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-InP.py

@@ -0,0 +1,31 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+state_dict = load_state_dict("models/train/Wan2.1-Fun-V1.1-14B-InP_full/epoch-1.safetensors")
+pipe.dit.load_state_dict(state_dict)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0], end_image=video[80],
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-14B-InP.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B-Preview.py

@@ -0,0 +1,30 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="iic/VACE-Wan2.1-1.3B-Preview", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="iic/VACE-Wan2.1-1.3B-Preview", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="iic/VACE-Wan2.1-1.3B-Preview", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+state_dict = load_state_dict("models/train/Wan2.1-VACE-1.3B-Preview_full/epoch-1.safetensors")
+pipe.vace.load_state_dict(state_dict)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1_softedge.mp4", height=480, width=832)
+video = [video[i] for i in range(49)]
+reference_image = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)[0]
+
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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",
+    vace_video=video, vace_reference_image=reference_image, num_frames=49,
+    seed=1, tiled=True
+)
+save_video(video, "video_Wan2.1-VACE-1.3B-Preview.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B.py

@@ -0,0 +1,30 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+state_dict = load_state_dict("models/train/Wan2.1-VACE-1.3B_full/epoch-1.safetensors")
+pipe.vace.load_state_dict(state_dict)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1_softedge.mp4", height=480, width=832)
+video = [video[i] for i in range(49)]
+reference_image = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)[0]
+
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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",
+    vace_video=video, vace_reference_image=reference_image, num_frames=49,
+    seed=1, tiled=True
+)
+save_video(video, "video_Wan2.1-VACE-1.3B.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_full/Wan2.1-VACE-14B.py

@@ -0,0 +1,30 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+state_dict = load_state_dict("models/train/Wan2.1-VACE-14B_full/epoch-1.safetensors")
+pipe.vace.load_state_dict(state_dict)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1_softedge.mp4", height=480, width=832)
+video = [video[i] for i in range(17)]
+reference_image = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)[0]
+
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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",
+    vace_video=video, vace_reference_image=reference_image, num_frames=17,
+    seed=1, tiled=True
+)
+save_video(video, "video_Wan2.1-VACE-14B.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py

@@ -0,0 +1,31 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+pipe.load_lora(pipe.dit, "models/train/Wan2.1-Fun-V1.1-1.3B-Control-Camera_lora/epoch-4.safetensors", alpha=1)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0],
+    camera_control_direction="Left", camera_control_speed=0.0,
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-1.3B-Control-Camera.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-InP.py

@@ -0,0 +1,30 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-1.3B-InP", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+pipe.load_lora(pipe.dit, "models/train/Wan2.1-Fun-V1.1-1.3B-InP_lora/epoch-4.safetensors", alpha=1)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0], end_image=video[80],
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-1.3B-InP.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control-Camera.py

@@ -0,0 +1,31 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData, load_state_dict
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control-Camera", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+pipe.load_lora(pipe.dit, "models/train/Wan2.1-Fun-V1.1-1.3B-Control-Camera_lora/epoch-4.safetensors", alpha=1)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0],
+    camera_control_direction="Left", camera_control_speed=0.0,
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-14B-Control-Camera.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-InP.py

@@ -0,0 +1,30 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-InP", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+    ],
+)
+pipe.load_lora(pipe.dit, "models/train/Wan2.1-Fun-V1.1-14B-InP_lora/epoch-4.safetensors", alpha=1)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)
+
+# First and last frame to video
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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=video[0], end_image=video[80],
+    seed=0, tiled=True
+)
+save_video(video, "video_Wan2.1-Fun-V1.1-14B-InP.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B-Preview.py

@@ -0,0 +1,29 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="iic/VACE-Wan2.1-1.3B-Preview", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="iic/VACE-Wan2.1-1.3B-Preview", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="iic/VACE-Wan2.1-1.3B-Preview", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+pipe.load_lora(pipe.vace, "models/train/Wan2.1-VACE-1.3B-Preview_lora/epoch-4.safetensors", alpha=1)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1_softedge.mp4", height=480, width=832)
+video = [video[i] for i in range(49)]
+reference_image = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)[0]
+
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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",
+    vace_video=video, vace_reference_image=reference_image, num_frames=49,
+    seed=1, tiled=True
+)
+save_video(video, "video_Wan2.1-VACE-1.3B-Preview.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B.py

@@ -0,0 +1,29 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-1.3B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+pipe.load_lora(pipe.vace, "models/train/Wan2.1-VACE-1.3B_lora/epoch-4.safetensors", alpha=1)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1_softedge.mp4", height=480, width=832)
+video = [video[i] for i in range(49)]
+reference_image = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)[0]
+
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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",
+    vace_video=video, vace_reference_image=reference_image, num_frames=49,
+    seed=1, tiled=True
+)
+save_video(video, "video_Wan2.1-VACE-1.3B.mp4", fps=15, quality=5)

examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-14B.py

@@ -0,0 +1,29 @@
+import torch
+from PIL import Image
+from diffsynth import save_video, VideoData
+from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
+
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-VACE-14B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
+    ],
+)
+pipe.load_lora(pipe.vace, "models/train/Wan2.1-VACE-14B_lora/epoch-4.safetensors", alpha=1)
+pipe.enable_vram_management()
+
+video = VideoData("data/example_video_dataset/video1_softedge.mp4", height=480, width=832)
+video = [video[i] for i in range(17)]
+reference_image = VideoData("data/example_video_dataset/video1.mp4", height=480, width=832)[0]
+
+video = pipe(
+    prompt="from sunset to night, a small town, light, house, river",
+    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",
+    vace_video=video, vace_reference_image=reference_image, num_frames=17,
+    seed=1, tiled=True
+)
+save_video(video, "video_Wan2.1-VACE-14B.mp4", fps=15, quality=5)

examples/wanvideo/wan_14b_text_to_video_usp.py

@@ -1,58 +0,0 @@
-import torch
-from diffsynth import ModelManager, WanVideoPipeline, save_video, VideoData
-from modelscope import snapshot_download
-import torch.distributed as dist
-
-
-# Download models
-snapshot_download("Wan-AI/Wan2.1-T2V-14B", local_dir="models/Wan-AI/Wan2.1-T2V-14B")
-
-# Load models
-model_manager = ModelManager(device="cpu")
-model_manager.load_models(
-    [
-        [
-            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00001-of-00006.safetensors",
-            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00002-of-00006.safetensors",
-            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00003-of-00006.safetensors",
-            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00004-of-00006.safetensors",
-            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00005-of-00006.safetensors",
-            "models/Wan-AI/Wan2.1-T2V-14B/diffusion_pytorch_model-00006-of-00006.safetensors",
-        ],
-        "models/Wan-AI/Wan2.1-T2V-14B/models_t5_umt5-xxl-enc-bf16.pth",
-        "models/Wan-AI/Wan2.1-T2V-14B/Wan2.1_VAE.pth",
-    ],
-    torch_dtype=torch.float8_e4m3fn, # You can set `torch_dtype=torch.bfloat16` to disable FP8 quantization.
-)
-
-dist.init_process_group(
-    backend="nccl",
-    init_method="env://",
-)
-from xfuser.core.distributed import (initialize_model_parallel,
-                                     init_distributed_environment)
-init_distributed_environment(
-    rank=dist.get_rank(), world_size=dist.get_world_size())
-
-initialize_model_parallel(
-    sequence_parallel_degree=dist.get_world_size(),
-    ring_degree=1,
-    ulysses_degree=dist.get_world_size(),
-)
-torch.cuda.set_device(dist.get_rank())
-
-pipe = WanVideoPipeline.from_model_manager(model_manager, 
-                                           torch_dtype=torch.bfloat16, 
-                                           device=f"cuda:{dist.get_rank()}", 
-                                           use_usp=True if dist.get_world_size() > 1 else False)
-pipe.enable_vram_management(num_persistent_param_in_dit=None) # You can set `num_persistent_param_in_dit` to a small number to reduce VRAM required.
-
-# Text-to-video
-video = pipe(
-    prompt="一名宇航员身穿太空服，面朝镜头骑着一匹机械马在火星表面驰骋。红色的荒凉地表延伸至远方，点缀着巨大的陨石坑和奇特的岩石结构。机械马的步伐稳健，扬起微弱的尘埃，展现出未来科技与原始探索的完美结合。宇航员手持操控装置，目光坚定，仿佛正在开辟人类的新疆域。背景是深邃的宇宙和蔚蓝的地球，画面既科幻又充满希望，让人不禁畅想未来的星际生活。",
-    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",
-    num_inference_steps=50,
-    seed=0, tiled=True
-)
-if dist.get_rank() == 0:
-    save_video(video, "video1.mp4", fps=25, quality=5)