Checklist
Describe the Bug
After upgraded to ComfyUI-nunchaku 1.0.1 and pytorch to 2.10, I cannot use nunchaku anymore. The install_wheel.json cannot find the correct whl like this:
❌ ERROR:
Failed to install from all available sources.
Last error: Installation failed for https://github.com/nunchaku-tech/nunchaku/releases/download/v1.0.1dev20250930/nunchaku-1.0.1.dev20250930+torch2.9-cp312-cp312-win_amd64.whl. Error: HTTP Error 404: Not Found
So I manually downloaded nunchaku-1.0.1.dev20250930+torch2.10-cp312-cp312-win_amd64.whl and nunchaku-1.0.1+torch2.10-cp312-cp312-win_amd64.whl and tried to install but neither of them works, it shows this:
======================================== ComfyUI-nunchaku Initialization ========================================
Nunchaku version: 1.0.1
ComfyUI-nunchaku version: 1.0.1
ComfyUI-nunchaku 1.0.1 is not compatible with nunchaku 1.0.1. Please update nunchaku to a supported version in ['v1.0.0'].
Environment
My environment:
** ComfyUI startup time: 2025-10-12 15:54:24.988
** Platform: Windows
** Python version: 3.12.7 | packaged by Anaconda, Inc. | (main, Oct 4 2024, 13:17:27) [MSC v.1929 64 bit (AMD64)]
** Python executable: C:\Users\Danie\anaconda3\python.exe
** ComfyUI Path: V:\ComfyUI
** ComfyUI Base Folder Path: V:\ComfyUI
** User directory: \\DSM_10TB\home\ComfyUI\ComfyUI\user
** ComfyUI-Manager config path: \\DSM_10TB\home\ComfyUI\ComfyUI\user\default\ComfyUI-Manager\config.ini
** Log path: \\DSM_10TB\home\ComfyUI\ComfyUI\user\comfyui.log
...
Total VRAM 16303 MB, total RAM 95887 MB
pytorch version: 2.10.0.dev20251010+cu130
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce RTX 5070 Ti : cudaMallocAsync
Using pytorch attention
Python version: 3.12.7 | packaged by Anaconda, Inc. | (main, Oct 4 2024, 13:17:27) [MSC v.1929 64 bit (AMD64)]
ComfyUI version: 0.3.64
ComfyUI frontend version: 1.27.10
CUDA:
PS C:\Users\Danie> nvidia-smi
Sun Oct 12 16:18:57 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 581.29 Driver Version: 581.29 CUDA Version: 13.0 |
+-----------------------------------------+------------------------+----------------------+
| GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 5070 Ti WDDM | 00000000:01:00.0 Off | N/A |
| 0% 39C P8 16W / 300W | 222MiB / 16303MiB | 0% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 1328 C ...rs\Danie\anaconda3\python.exe N/A |
+-----------------------------------------------------------------------------------------+
Reproduction Steps
Already tried nunchaku-1.0.1.dev20250930+torch2.10-cp312-cp312-win_amd64.whl and nunchaku-1.0.1+torch2.10-cp312-cp312-win_amd64.whl but none of them works.
Checklist
Describe the Bug
After upgraded to ComfyUI-nunchaku 1.0.1 and pytorch to 2.10, I cannot use nunchaku anymore. The install_wheel.json cannot find the correct whl like this:
So I manually downloaded nunchaku-1.0.1.dev20250930+torch2.10-cp312-cp312-win_amd64.whl and nunchaku-1.0.1+torch2.10-cp312-cp312-win_amd64.whl and tried to install but neither of them works, it shows this:
Environment
My environment:
CUDA:
Reproduction Steps
Already tried nunchaku-1.0.1.dev20250930+torch2.10-cp312-cp312-win_amd64.whl and nunchaku-1.0.1+torch2.10-cp312-cp312-win_amd64.whl but none of them works.