Wan2.1图生视频,Comfyui一键生成涩图，换脸高清修復一体包

a931386

zhxbkofkyo 发表于 2025-7-2 19:56
大佬，WanVideo Enhance A Video (native)
Supported on ComfyICUSupported on ComfyICU
这个红色框框。 ...

這邊看是不是改成這個

xiangsir

"Wan2.1图生视频整合包"运行后提示很多节点都缺失也

qq312088415

6666666

xiangsir

有没有大佬帮忙看看是哪里出了问题，感谢！
Requested to load WanVAE
loaded completely 3313.1971015930176 242.02829551696777 True
  0%|                                                                                           | 0/80 [00:00<?, ?it/s]D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\raft\raft.py:121: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(enabled=self.args.mixed_precision):
D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\raft\raft.py:132: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(enabled=self.args.mixed_precision):
D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\raft\raft.py:149: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(enabled=self.args.mixed_precision):
  0%|                                                                                           | 0/80 [00:00<?, ?it/s]
!!! Exception during processing !!! 'CUDA_HOME' not set, unable to find cuda-toolkit installation.
Traceback (most recent call last):
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\modules\softsplat.py", line 267, in cuda_launch
    os.environ.setdefault("CUDA_HOME", cupy.cuda.get_cuda_path())
  File "<frozen os>", line 753, in setdefault
  File "<frozen os>", line 719, in __setitem__
  File "<frozen os>", line 779, in check_str
TypeError: str expected, not NoneType

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\execution.py", line 347, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\execution.py", line 222, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\execution.py", line 194, in _map_node_over_list
    process_inputs(input_dict, i)
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\execution.py", line 183, in process_inputs
    results.append(getattr(obj, func)(**inputs))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\nodes.py", line 213, in interpolate
    all_outputs = gimmvfi_model(xs, coord_inputs, t=timesteps, ds_factor=ds_factor)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\gimmvfi_r.py", line 356, in forward
    normal_inr_flows = self.predict_flow(normal_flows, coord, t, flows)
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\gimmvfi_r.py", line 175, in predict_flow
    tmp_pixel_latent_0 = softsplat(
                         ^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 838, in _fn
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\modules\softsplat.py", line 321, in softsplat
    tenOut = softsplat_func.apply(tenIn, tenFlow)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\python_embeded\Lib\site-packages\torch\autograd\function.py", line 575, in apply
    return super().apply(*args, **kwargs)  # type: ignore[misc]
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\python_embeded\Lib\site-packages\torch\amp\autocast_mode.py", line 521, in decorate_fwd
    return fwd(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\modules\softsplat.py", line 370, in forward
    cuda_launch(
  File "cupy\\_util.pyx", line 64, in cupy._util.memoize.decorator.ret
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\python_embeded\Lib\site-packages\torch\_dynamo\eval_frame.py", line 838, in _fn
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "D:\AI-tools\Wan2.1-i2v\ComfyUI_windows_WanVideo\ComfyUI_windows_WanVideo\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\modules\softsplat.py", line 270, in cuda_launch
    raise RuntimeError("'CUDA_HOME' not set, unable to find cuda-toolkit installation.")
RuntimeError: 'CUDA_HOME' not set, unable to find cuda-toolkit installation.

a931386

xiangsir 发表于 2025-7-4 16:51
有没有大佬帮忙看看是哪里出了问题，感谢！
Requested to load WanVAE
loaded completely 3313.19710159301 ...

CUDA_HOME 没有设置环境变量

要安裝顯卡對應的CUDA Toolkit

a931386

xiangsir 发表于 2025-7-4 14:31
"Wan2.1图生视频整合包"运行后提示很多节点都缺失也

我有準備整合包和刪除模型只帶節點的Comfyui以防爆錯,兩種可以擇一

xiangsir

a931386 发表于 2025-7-4 21:10
CUDA_HOME 没有设置环境变量

要安裝顯卡對應的CUDA Toolkit

是的，我以为我安装了的，结果发现那是之前玩儿Linux的时候装的，记忆错乱了

xiangsir

a931386 发表于 2025-7-4 21:11
我有準備整合包和刪除模型只帶節點的Comfyui以防爆錯,兩種可以擇一

这个好像是因为浏览器缓存的原因，后面就没问题了，感谢！

1235Aj

你好，大佬，我想问一下，comfyui怎么在同一个页面切换不同的工作流，我打开一个工作流里面有很多其他不同的工作流，但是只能用一个其他的都是灰色的。

a931386

1235Aj 发表于 2025-7-29 12:39
你好，大佬，我想问一下，comfyui怎么在同一个页面切换不同的工作流，我打开一个工作流里面有很多其他不同 ...

快捷 CTRL+B 關閉節點可以實現同一頁面塞進不同工作流