Help with Runtime Error

[w9ran]

First, confirm

• I have read the instruction carefully
• I have searched the existing issues
• I have updated the extension to the latest version

What happened?

When attempting to create my first short video (4 second video) using Video Helper Suite. The workflow starts processing as normal but after a few minutes the following error occurs (tail of error log)

File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 298, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 286, in process_inputs
result = f(**inputs)
File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor\nodes.py", line 1470, in execute
result[image_index] = pasting * paste_mask + result[image_index] * (1. - paste_mask)
~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
RuntimeError: The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension 2

2025-11-21T17:56:56.780104 - Prompt executed in 375.03 seconds

I'd appreciate suggestions about how to resolve this as the error message is rather cryptic.
The workflow seems to execute as it should be the console shows repetitive errors:
"cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")
I don't know if these errors are the cause of the Runtime Error or not. Google turned up other examples of this error but no obvious solutions.

ReActor works fine on images, fast and no errors.

The same issue was opened on Comfyui but did not get any responses, probably because the error is from ReActorMaskHelper, thus is not a Comfyui issue.

Thanks for any help.

Steps to reproduce the problem

Your workflow

ReactModel2.json

Sysinfo

Windows 11 Pro, Firefox 145.0
AMD Ryzen 7 5800XT 8-Core Processor 3.80 GHz 64GB
RTX-3090 24 GB
This is a clean installation of Comfyui with the workflow from https://www.youtube.com/watch?v=cFrDYl1FJwc
The only other custom node is Video Helper Suite for convert the video to images.

Relevant console log

Since my log is too large to fit this field I have redacted about 140 copies of the "cublasLt64_12.dll is missing" error, leaving about 10 in the version below.  I don't know if this is the cause of the runtime error or if so, how to fix it.
---

C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build
[START] Security scan
[DONE] Security scan
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2025-11-21 20:31:22.646
** Platform: Windows
** Python version: 3.13.9 (tags/v3.13.9:8183fa5, Oct 14 2025, 14:09:13) [MSC v.1944 64 bit (AMD64)]
** Python executable: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\python.exe
** ComfyUI Path: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
** ComfyUI Base Folder Path: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
** User directory: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\user
** ComfyUI-Manager config path: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Manager\config.ini
** Log path: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\user\comfyui.log

Prestartup times for custom nodes:
   2.3 seconds: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager

Checkpoint files will always be loaded safely.
Total VRAM 24564 MB, total RAM 65462 MB
pytorch version: 2.9.0+cu130
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 3090 Ti : cudaMallocAsync
working around nvidia conv3d memory bug.
Using pytorch attention
Python version: 3.13.9 (tags/v3.13.9:8183fa5, Oct 14 2025, 14:09:13) [MSC v.1944 64 bit (AMD64)]
ComfyUI version: 0.3.68
ComfyUI frontend version: 1.28.8
[Prompt Server] web root: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\comfyui_frontend_package\static
### Loading: ComfyUI-Manager (V3.37.1)
[ComfyUI-Manager] network_mode: public
### ComfyUI Revision: 150 [265adad8] *DETACHED | Released on '2025-11-04'
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json

[ReActor] - STATUS - Running v0.6.2-a4 in ComfyUI
Torch version: 2.9.0+cu130

Import times for custom nodes:
   0.0 seconds: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py
   0.0 seconds: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-videohelpersuite
   0.1 seconds: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager
   1.5 seconds: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor

Context impl SQLiteImpl.
Will assume non-transactional DDL.
No target revision found.
Starting server

To see the GUI go to: http://127.0.0.1:8188
FETCH ComfyRegistry Data: 5/107
[DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
[DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/groupNode.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
[DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/widgetInputs.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
[DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/buttonGroup.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
[DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/button.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version.
FETCH ComfyRegistry Data: 10/107
FETCH ComfyRegistry Data: 15/107
FETCH ComfyRegistry Data: 20/107
FETCH ComfyRegistry Data: 25/107
FETCH ComfyRegistry Data: 30/107
FETCH ComfyRegistry Data: 35/107
FETCH ComfyRegistry Data: 40/107
FETCH ComfyRegistry Data: 45/107
FETCH ComfyRegistry Data: 50/107
FETCH ComfyRegistry Data: 55/107
FETCH ComfyRegistry Data: 60/107
FETCH ComfyRegistry Data: 65/107
got prompt
2025-11-21 20:32:12.6977939 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

FETCH ComfyRegistry Data: 70/107
2025-11-21 20:32:13.1692053 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

duplicated model task type, ignore: C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\insightface\models\buffalo_l\2d106det.onnx landmark
2025-11-21 20:32:13.2732100 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

2025-11-21 20:32:13.3361393 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

2025-11-21 20:32:13.4246563 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:14 - STATUS - Checking for any unsafe content...
[ReActor] 20:32:17 - STATUS - Working: source face index [0], target face index [0]
[ReActor] 20:32:17 - STATUS - Using Loaded Source Face Model...
[ReActor] 20:32:17 - STATUS - Analyzing Target Image...
FETCH ComfyRegistry Data: 75/107
FETCH ComfyRegistry Data: 80/107
FETCH ComfyRegistry Data: 85/107
FETCH ComfyRegistry Data: 90/107
FETCH ComfyRegistry Data: 95/107
FETCH ComfyRegistry Data: 100/107
2025-11-21 20:32:37.5754845 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:38 - STATUS - Swapping...
[ReActor] 20:32:38 - STATUS - Face Boost is enabled
[ReActor] 20:32:39 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:39.4084694 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

FETCH ComfyRegistry Data: 105/107
[ReActor] 20:32:41 - STATUS - Face Boost is enabled
FETCH ComfyRegistry Data [DONE]
[ReActor] 20:32:41 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:41.7872813 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json [DONE]
[ComfyUI-Manager] All startup tasks have been completed.
[ReActor] 20:32:43 - STATUS - Face Boost is enabled
[ReActor] 20:32:44 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:44.2345990 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:45 - STATUS - Face Boost is enabled
[ReActor] 20:32:46 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:46.6367129 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:48 - STATUS - Face Boost is enabled
[ReActor] 20:32:48 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:49.0736489 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:50 - STATUS - Face Boost is enabled
[ReActor] 20:32:51 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:51.4984886 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:53 - STATUS - Face Boost is enabled
[ReActor] 20:32:53 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:53.9222812 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:55 - STATUS - Face Boost is enabled
[ReActor] 20:32:56 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:56.3497482 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:32:58 - STATUS - Face Boost is enabled
[ReActor] 20:32:58 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:32:58.7749362 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:33:00 - STATUS - Face Boost is enabled
[ReActor] 20:33:01 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-21 20:33:01.2216484 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[:ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

[ReActor] 20:38:28 - STATUS - --Done!--
[ReActor] 20:38:29 - STATUS - Restoring with codeformer-v0.1.0.pth | Face Size is set to 512


0: 640x480 1 face, 50.5ms
0: 640x480 1 face, 50.5ms
Speed: 12.7ms preprocess, 50.5ms inference, 1.8ms postprocess per image at shape (1, 3, 640, 480)
Speed: 12.7ms preprocess, 50.5ms inference, 1.8ms postprocess per image at shape (1, 3, 640, 480)
Moving to GPU... OK
!!! Exception during processing !!! The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension 2
Traceback (most recent call last):
  File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 510, in execute
    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)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 324, in get_output_data
    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)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 298, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 286, in process_inputs
    result = f(**inputs)
  File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor\nodes.py", line 1470, in execute
    result[image_index] = pasting * paste_mask + result[image_index] * (1. - paste_mask)
                                                 ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
RuntimeError: The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension 2

Prompt executed in 454.85 seconds

Additional information

Workflow works fine with Source Image and with face model.

[Buumcode]

[ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

You should fix this error.

You don't have properly installed onnxruntime-gpu.

Try this (in your python embedded):
Check your Reactor uses GPU -- not CPU!

pip uninstall onnxruntime
pip uninstall onnxruntime-gpu

pip install onnxruntime-gpu

[w9ran]

Thanks very much! No doubt this is my problem because I heard the fans speed up and the CPU was working hard.
But I've run pip list in all the onnx directories and all it tells me is about pip, nothing about onnx:

\onnx> pip list
Package Version

---

pip 24.3.1

I do have the following directories under ComfyUI/windows_portable\python_embeded\Lib\site-packages:
onnx
onnx-1.19.1dist-info
onnxruntime_gpu-1.23.2.dist-info

My search tool finds multiple hits for "onnxruntime" including in the Comfyui directory I'm working from, but does not find "onnxruntime-gpu" anywhere. So it needs to be installed using pip install onnxruntime.gpu, correct?

It would be very helpful if you could tell me exactly WHERE to run this command. I'm afraid of messing something up!
If I open a terminal in \Lib\site-packages it should end up installing onnxruntime.gpu at a peer level to the above three onnx-related files - is that where it should be?

Thanks again with this issue.

[Buumcode]

You better ask from ChatGPT for quick help (ask in your native language) - https://gpt-chatbot.ru/gpt-5

I'm not sure about Portable python - what command do you need for pip list, try this (launch cmd.exe from your ComfyUI folder - click mouse on adress bar, than type 'cmd' - press Enter):
INFO: python.exe -m <--- prefix for portable python:

python.exe -m pip list

And after that, do that (three commands, step-by-step):

python.exe -m pip uninstall onnxruntime
python.exe -m pip uninstall onnxruntime-gpu

python.exe -m pip install onnxruntime-gpu

[w9ran]

Thanks again. The ChatGPT instructions seemed quite clear: "You do NOT install it system-wide — you install it inside ComfyUI’s embedded Python."
So I did that after closing Comfyui and confirmed it was installed by running the following test that was provided:

PS C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded> python -c "import onnxruntime as ort; print(ort.get_device())"
GPU
PS C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded>

The response "GPU" indicates that it is installed.
But when I try to run the workflow using a short video to generate the images, it still fails with the same runtime error: "The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension"
Looking through the log file it's no different than before, with an instance of this error for (I assume) each frame:

[ReActor] 14:04:01 - STATUS - Face Boost is enabled
[ReActor] 14:04:01 - STATUS - Boosting the Face with GPEN-BFR-512.onnx | Face Size is set to 512 with Scale Factor = 4.0 and 'Nearest' interpolation
2025-11-22 14:04:02.0294144 [E:onnxruntime:Default, provider_bridge_ort.cc:2251 onnxruntime::TryGetProviderInfo_CUDA] E:_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1844 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")

To be sure, I re-deleted both onnxruntime and onnxruntime-gpu and reinstalled the latter. Pip list then showed it to be present:

PS C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded> pip list
Package Version

---

coloredlogs 15.0.1
flatbuffers 25.9.23
humanfriendly 10.0
mpmath 1.3.0
numpy 2.3.5
onnxruntime-gpu 1.23.2
packaging 25.0
pip 25.3
protobuf 6.33.1
pyreadline3 3.5.4
sympy 1.14.0

So I restarted my workflow but once again it failed with the same "module could not be found" message each frame and with the following at the tail of the console:
File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 298, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 286, in process_inputs
result = f(**inputs)
File "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor\nodes.py", line 1470, in execute
result[image_index] = pasting * paste_mask + result[image_index] * (1. - paste_mask)
~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
RuntimeError: The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension 2

Prompt executed in 246.57 seconds

For most of the time the CPU was running 70%+ and the GPU was running at 6-8% utilization. Seems to me I need "cublasLt64_12.dll"

Thanks in advance for any suggestions.

[Buumcode]

Post this error to ChatGPT and ask what to do to fix that:
"[ONNXRuntimeError] : 1 : FAIL : Error loading "C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cublasLt64_12.dll" which is missing. (Error 126: "The specified module could not be found.")"

Do you have this file - C:\Users\Administrator\Downloads\Exp_ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll ??