- **Exception Message:** cuDNN Frontend error: CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM

[AS1ING]

XLabs-AI/flux-ip-adapter-v2

ComfyUI Error Report

Error Details

• Node Type: XlabsSampler
• Exception Type: RuntimeError
• Exception Message: cuDNN Frontend error: CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM

Stack Trace

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\execution.py", line 323, 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 "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\execution.py", line 198, 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 "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\execution.py", line 169, in _map_node_over_list
    process_inputs(input_dict, i)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\execution.py", line 158, in process_inputs
    results.append(getattr(obj, func)(**inputs))

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\custom_nodes\x-flux-comfyui\nodes.py", line 411, in sampling
    x = denoise(

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\custom_nodes\x-flux-comfyui\sampling.py", line 193, in denoise
    pred = model_forward(

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\custom_nodes\x-flux-comfyui\sampling.py", line 51, in model_forward
    img, txt = block(img=img, txt=txt, vec=vec, pe=pe)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\venv\lib\site-packages\torch\nn\modules\module.py", line 1736, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\venv\lib\site-packages\torch\nn\modules\module.py", line 1747, in _call_impl
    return forward_call(*args, **kwargs)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\custom_nodes\x-flux-comfyui\xflux\src\flux\modules\layers.py", line 297, in forward
    return self.processor(self, img, txt, vec, pe)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\venv\lib\site-packages\torch\nn\modules\module.py", line 1736, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\venv\lib\site-packages\torch\nn\modules\module.py", line 1747, in _call_impl
    return forward_call(*args, **kwargs)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\custom_nodes\x-flux-comfyui\layers.py", line 388, in forward
    img = self.shift_ip(img_q, attn, img)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\custom_nodes\x-flux-comfyui\layers.py", line 313, in shift_ip
    x += torch.mean(block(img_qkv, attn), dim=0, keepdim=True)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\venv\lib\site-packages\torch\nn\modules\module.py", line 1736, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\venv\lib\site-packages\torch\nn\modules\module.py", line 1747, in _call_impl
    return forward_call(*args, **kwargs)

  File "F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\custom_nodes\x-flux-comfyui\layers.py", line 265, in forward
    ip_attention = F.scaled_dot_product_attention(

System Information

• ComfyUI Version: v0.2.4-12-g669d9e4
• Arguments: F:\StabilityMatrix-win-x64\Data\Packages\ComfyUI Test\main.py --preview-method auto --auto-launch --fast --reserve-vram 0.8
• OS: nt
• Python Version: 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]
• Embedded Python: false
• PyTorch Version: 2.5.0+cu124

Devices

• Name: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
  • Type: cuda
  • VRAM Total: 25756696576
  • VRAM Free: 2406419424
  • Torch VRAM Total: 20434649088
  • Torch VRAM Free: 34708448

[caslix]

Hello! I have the same problem.

[rgfx]

This error occurs on the new workspace that was given. If I used the old one, and switch the modal to the new one it works.

[christming]

This error occurs on the new workspace that was given. If I used the old one, and switch the modal to the new one it works.

Hello~ how do you fix this bug? Can you give some more details here? I meet the same error with use the new XLabs-AI/flux-ip-adapter-v2

[xlibfly]

I have the same problem.

[lior007]

same

[Orenji-Tangerine]

I have the same problem.

[xieyao2]

same problem.rgfx how did u fix it?

[xieyao2]

XLabs-AI/flux-ip-adapter-v2
14700kf
4060ti 16gb
ram 64gb
torch2.5.0 + cuda12.4+xformers0.0.28.post2

[xieyao2]

when i use one image,the workflow is ok.