Wan2.2-Dasiwa - 专注动漫及艺术类视频生成 支持NSFW 一键整合下载

无言***

Wan2.2-Dasiwa - 专注动漫及艺术类视频生成 支持NSFW 一键整合下载

Wan2.2-Dasiwa （模型如其名，大丝袜），这是是一个基于Wan2.2的优化版高级视频生成模型，它就像一个“超级会拍艺术类短视频的AI导演”。你给它一张图片，它就能自动生成一段动作自然、细节丰富、很感性的短视频。特别擅长把静态图片变成动态的、诱惑力强的艺术视频。
Wan2.2-Dasiwa 默认已包含 NSFW 功能，能够生成高质量的 NSFW 主题视频，尤其是动漫类视频生成，无需额外的 LoRA。且经过多次优化改进，无论是生成视频质量还是动画效果，都优于同类开源模型。


主要特点

动作更流畅、自然：比很多同类模型动得更多、更真实（提升20-30%动作量），不会看起来僵硬或慢动作过多。
细节超强：即使分辨率不高，皮肤、眼睛、嘴唇、胸部、私密部位等都画得很精致，几乎全程保持一致（很少出现“视频里突然变丑”的情况）。
听提示词能力强：你写什么它就容易做出什么（prompt adherence高），包括各种姿势、表情、特殊效果。
很适合NSFW：内置了丰富的性感理解，能轻松生成诱惑、性感动作等，但也不会过度偏向某个极端风格。



应用领域

主要用于艺术视频生成（NSFW短视频、动态图、OnlyFans风格内容等）。
也支持普通SFW（非艺术类）内容，比如动画风、美女动态肖像、时尚动作等。
适合做办公室OL、丝袜诱惑、慢挑逗、性感舞蹈这类带故事感的短视频（跟你之前写的那个视频提示词风格非常搭）。




使用教程：（建议N卡，显存8G起，支持50系显卡）

整合包包含所需所有节点，下载主程序和模型（ComfyUI文件夹），解压主程序一键包，将ComfyUI文件夹移动到主程序目录下即可。

ComfyUI模式：
双击启动ComfyUI，进入WebUI后，点击左侧的 工作流程，选择对应的工作流，根据需要选择图生视频和首尾帧视频生成，左侧有开关，根据需要开启和关闭。
上传需要生成的图像，输入提示词，设置相关参数，运行即可。
实测8G显存可以运行，但需要开启共享显存，且运存需要32G起。

WebUI模式：
双击启动Webui，根据需要切换图生视频和首尾帧视频生成选项卡，上传图像，输入提示词，设置相关参数，运行生成即可。
模型切换：（包含高噪和低噪模型，模型手动下载后，保存到 ComfyUI\models\unet 目录下，运行WebUI，在模型切换下拉选择更新的模型）

默认开启Sage加速，如报错，可手动关闭加速。注. 开启sage_attention 加速需要本地安装VC编译环境，具体安装教程参考 此贴

提示词（prompt）写得越详细越好，比如加上“流畅运动、自然光影、高细节、电影感”等，能让效果更好。
一键包内包含 提示词生成技巧教程，教程来自B站UP Work-Fisher


软件目录结构：

📂 ComfyUI/
├── 📂 models/
│   ├── 📂 unet/
│   │      └── Dasiwa_WAN22-I2V14BSynthseduction_q4High.gguf
│   │      └── Dasiwa_WAN22-I2V14BSynthseduction_q4Low.gguf
│   ├── 📂 text_encoders/
│   │      └── nsfw_wan_umt5-xxl_fp8_scaled.safetensors
│   └── 📂 vae/
│          └── wan_2.1_vae.safetensors
📂 deepface/
......



下载地址：
迅雷云盘：https://pan.xunlei.com/s/VOp5qP1f9G0LEce98jq71UUkA1?pwd=m7km

夸克网盘：

🔒付费内容

游客， 上上宾会员 可免费下载该资源，点此开通上上宾 免费下载全站99%的付费资源。或单独支付 50碎银 下载该资源

立即购买

百度网盘：

🔒付费内容

游客， 上上宾会员 可免费下载该资源，点此开通上上宾 免费下载全站99%的付费资源。或单独支付 66碎银 下载该资源

立即购买

lon***

你好，wan2.2和wan2gp哪个更好用？显卡是4060，8g显存

无言***

longli 发表于 2026-4-1 15:30
你好，wan2.2和wan2gp哪个更好用？显卡是4060，8g显存

wan2gp

lon***

无言以对 发表于 2026-4-1 15:32
wan2gp

好的，感谢解答

cc142***

用迅雷下载的ComfyUI文件夹，里面只有models\configs一个文件夹呢

无言***

cc142006596 发表于 2026-4-7 21:42
用迅雷下载的ComfyUI文件夹，里面只有models\configs一个文件夹呢

被吞了，已补

31535***

大佬，求助，请教个问题，执行生成时报错，第一次是提交任务失败，说是加载vae问题，手动选择了pixel_space这个之后解决了，问题1：后面又遇到问题，查资料好像说是vae默认用的wan2.1，不兼容，要用wan2.2，问题2：我下载了wan2.2放在新建的vae目录下，vae加载器选择这个又报错了，好像还是说这个vae加载用的默认就是2.1的，新建vae加载器其实也是2.1的，和wan2.2的模型不兼容，不知道我前面解决的是否正确，所有报错都贴出来了，从最初的报错开始：报错1：
“[14:55:38] [ComfyUI] got prompt
[14:55:38] [ComfyUI] Failed to validate prompt for output 51:
[14:55:38] [ComfyUI] * VAELoader 18:
[14:55:38] [ComfyUI] - Value not in list: vae_name: 'wan_2.1_vae.safetensors' not in ['pixel_space']
[14:55:38] [ComfyUI] Output will be ignored
[14:55:38] [ComfyUI] Failed to validate prompt for output 56:
[14:55:38] [ComfyUI] Output will be ignored
[14:55:38] [ComfyUI] Failed to validate prompt for output 48:
[14:55:38] [ComfyUI] Output will be ignored
[14:55:38] [ComfyUI] invalid prompt: {'type': 'prompt_outputs_failed_validation', 'message': 'Prompt outputs failed validation', 'details': '', 'extra_info': {}}
[14:55:38] [Gradio] Traceback (most recent call last):
[14:55:38] [Gradio] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\deepface\Lib\site-packages\gradio\queueing.py", line 759, in process_events
[14:55:38] [Gradio] response = await route_utils.call_process_api(

[14:55:38] [Gradio] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\deepface\Lib\site-packages\gradio\blocks.py", line 1623, in call_function
[14:55:38] [Gradio] prediction = await anyio.to_thread.run_sync(  # type: ignore

[14:55:38] [Gradio] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\deepface\Lib\site-packages\anyio\to_thread.py", line 56, in run_sync
[14:55:38] [Gradio] return await get_async_backend().run_sync_in_worker_thread(
[14:55:38] [Gradio] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[14:55:38] [Gradio] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\deepface\Lib\site-packages\anyio\_backends\_asyncio.py", line 2485, in run_sync_in_worker_thread
[14:55:38] [Gradio] return await future
[14:55:38] [Gradio] ^^^^^^^^^^^^
[14:55:38] [Gradio] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\deepface\Lib\site-packages\anyio\_backends\_asyncio.py", line 976, in run
[14:55:38] [Gradio] result = context.run(func, *args)
[14:55:38] [Gradio] ^^^^^^^^^^^^^^^^^^^^^^^^
[14:55:38] [Gradio] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\deepface\Lib\site-packages\gradio\utils.py", line 915, in wrapper
[14:55:38] [Gradio] response = f(*args, **kwargs)
[14:55:38] [Gradio] ^^^^^^^^^^^^^^^^^^
[14:55:38] [Gradio] File "<frozen app>", line 760, in generate_i2v
[14:55:38] [Gradio] gradio.exceptions.Error: '提交任务失败'
”

报错2：
生成报错，“[15:28:29] [ComfyUI] got prompt
[15:28:29] [ComfyUI] Requested to load PixelspaceConversionVAE
[15:28:29] [ComfyUI] loaded completely 7591.3107421875 1.9073486328125e-06 True
[15:28:30] [ComfyUI] 0 models unloaded.
[15:28:30] [ComfyUI] loaded partially 128.0 127.998046875 0
[15:28:30] [ComfyUI] # 😺dzNodes: LayerStyle -> ImageScaleByAspectRatio V2 Processed 1 image(s).
[15:28:30] [ComfyUI] Initializing block swap:   0%|          | 0/40 [00:00<?, ?it/s]
[15:28:30] [ComfyUI] Initializing block swap: 100%|██████████| 40/40 [00:00<00:00, 889.53it/s]
[15:28:30] [ComfyUI] !!! Exception during processing !!! The size of tensor a (81) must match the size of tensor b (16) at non-singleton dimension 1
[15:28:30] [ComfyUI] Traceback (most recent call last):
[15:28:30] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\execution.py", line 498, in execute
[15:28:30] [ComfyUI] output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
[15:28:30] [ComfyUI]

[15:28:30] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\execution.py", line 316, in get_output_data
[15:28:30] [ComfyUI] return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
[15:28:30] [ComfyUI]

[15:28:30] [ComfyUI]
[15:28:30] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\comfy\latent_formats.py", line 453, in process_in
[15:28:30] [ComfyUI] return (latent - latents_mean) * self.scale_factor / latents_std
[15:28:30] [ComfyUI] ~~~~~~~^~~~~~~~~~~~~~
[15:28:30] [ComfyUI] RuntimeError: The size of tensor a (81) must match the size of tensor b (16) at non-singleton dimension 1
[15:28:30] [ComfyUI] Prompt executed in 1.33 seconds
[15:28:41] [ComfyUI] got prompt
[15:28:41] [ComfyUI] 0 models unloaded.
[15:28:41] [ComfyUI] loaded partially 128.0 127.998046875 0
[15:28:41] [ComfyUI] Initializing block swap:   0%|          | 0/40 [00:00<?, ?it/s]
[15:28:42] [ComfyUI] Initializing block swap: 100%|██████████| 40/40 [00:00<00:00, 865.38it/s]
[15:28:42] [ComfyUI] !!! Exception during processing !!! The size of tensor a (81) must match the size of tensor b (16) at non-singleton dimension 1
[15:28:42] [ComfyUI] Traceback (most recent call last):
[15:28:42] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\execution.py", line 498, in execute
[15:28:42] [ComfyUI] output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
[15:28:42] [ComfyUI]
[15:28:42] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\comfy\model_base.py", line 1125, in concat_cond
[15:28:42] [ComfyUI] image[:, i: i + latent_dim] = self.process_latent_in(image[:, i: i + latent_dim])
[15:28:42] [ComfyUI] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[15:28:42] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\comfy\model_base.py", line 317, in process_latent_in
[15:28:42] [ComfyUI] return self.latent_format.process_in(latent)
[15:28:42] [ComfyUI] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[15:28:42] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\comfy\latent_formats.py", line 453, in process_in
[15:28:42] [ComfyUI] return (latent - latents_mean) * self.scale_factor / latents_std
[15:28:42] [ComfyUI] ~~~~~~~^~~~~~~~~~~~~~
[15:28:42] [ComfyUI] RuntimeError: The size of tensor a (81) must match the size of tensor b (16) at non-singleton dimension 1
[15:28:42] [ComfyUI] Prompt executed in 0.43 seconds
”

报错3：
我在vae加载器中选择pixel_space还是报错，选择wan2.2vae可以，但是又出现以下报错“[15:46:05] [ComfyUI] got prompt

[15:46:35] [ComfyUI] Attempting to release mmap (468)
[15:46:46] [ComfyUI] !!! Exception during processing !!! The size of tensor a (81) must match the size of tensor b (16) at non-singleton dimension 1
[15:46:46] [ComfyUI] Traceback (most recent call last):
[15:46:46] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\execution.py", line 498, in execute
[15:46:46] [ComfyUI] output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
[15:46:46] [ComfyUI]
[15:46:46] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\execution.py", line 316, in get_output_data
[15:46:46] [ComfyUI] return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
[15:46:46] [ComfyUI]
[15:46:46] [ComfyUI] ~~~~~~~^~~~~~~~~~~~~~
[15:46:46] [ComfyUI] RuntimeError: The size of tensor a (81) must match the size of tensor b (16) at non-singleton dimension 1
[15:46:46] [ComfyUI] Prompt executed in 40.68 seconds
[15:47:09] [ComfyUI] got prompt
[15:47:09] [ComfyUI] Using pytorch attention in VAE
[15:47:09] [ComfyUI] Using pytorch attention in VAE
[15:47:10] [ComfyUI] VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
[15:47:11] [ComfyUI] Requested to load WanVAE
[15:47:11] [ComfyUI] loaded completely 4394.8172607421875 1344.0869674682617 True
[15:47:19] [ComfyUI] 0 models unloaded.

[15:47:20] [ComfyUI] !!! Exception during processing !!! Given groups=1, weight of size [5120, 36, 1, 2, 2], expected input[1, 64, 21, 128, 72] to have 36 channels, but got 64 channels instead
[15:47:20] [ComfyUI] Traceback (most recent call last):
[15:47:20] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\execution.py", line 498, in execute
[15:47:20] [ComfyUI] output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
[15:47:20] [ComfyUI]

[15:47:20] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\ComfyUI\comfy\ops.py", line 172, in _conv_forward
[15:47:20] [ComfyUI] return super()._conv_forward(input, weight, bias, *args, **kwargs)
[15:47:20] [ComfyUI] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[15:47:20] [ComfyUI] File "C:\AI\ComfyUI_Wan2.2-Dasiwa\deepface\Lib\site-packages\torch\nn\modules\conv.py", line 720, in _conv_forward
[15:47:20] [ComfyUI] return F.conv3d(
[15:47:20] [ComfyUI] ^^^^^^^^^
[15:47:20] [ComfyUI] RuntimeError: Given groups=1, weight of size [5120, 36, 1, 2, 2], expected input[1, 64, 21, 128, 72] to have 36 channels, but got 64 channels instead
[15:47:20] [ComfyUI] Prompt executed in 11.51 seconds
”

无言***

3153532099 发表于 2026-4-9 18:00
大佬，求助，请教个问题，执行生成时报错，第一次是提交任务失败，说是加载vae问题，手动选择了pixel_space ...

vae没问题，检查你的帧数设置和分辨率设置
帧数为16的倍数+1
用webui默认的参数生成试试

31535***

无言以对 发表于 2026-4-9 18:31
vae没问题，检查你的帧数设置和分辨率设置
帧数为16的倍数+1
用webui默认的参数生成试试

感谢大佬的回复，参数是用的默认的，帧数和分辨率都是。无意中，回头又看到第一个报错“Value not in list: vae_name: 'wan_2.1_vae.safetensors' not in ['pixel_space']”，我找了一个wan_2.1_vae.safetensors，把它放在vae目录下，vae加载器选择加载这个，结果好了。。。

1234***

3153532099 发表于 2026-4-9 19:05
感谢大佬的回复，参数是用的默认的，帧数和分辨率都是。无意中，回头又看到第一个报错“Value not in lis ...

在哪找的 wan_2.1_vae.safetensors ？？求问 我也这问题