# Unable to start comfyui cida ossie

**URL:** https://forum.comfy.org/t/unable-to-start-comfyui-cida-ossie/3822
**Category:** Windows
**Tags:** starting-comfyui, issue
**Created:** [September 26, 2025, 5:08pm UTC](https://forum.comfy.org/t/unable-to-start-comfyui-cida-ossie/3822 "2025-09-26T17:08:24Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![unrealsakura](https://avatars.discourse-cdn.com/v4/letter/u/a9adbd/32.png) [@unrealsakura](https://forum.comfy.org/u/unrealsakura)
#### Post date: [September 26, 2025, 5:08pm UTC](https://forum.comfy.org/t/unable-to-start-comfyui-cida-ossie/3822/1 "2025-09-26T17:08:24Z")

</div>

[2025-09-26 22:32:07.932] [error] C:\ComfyUI.venv\Lib\site-packages\torch\cuda\__init_\_.py:63: FutureWarning: The pynvml package is deprecated. Please install nvidia-ml-py instead. If you did not install pynvml directly, please report this to the maintainers of the package that installed pynvml for you.  
import pynvml # type: ignore[import]

[2025-09-26 22:32:10.937] [info] Checkpoint files will always be loaded safely.

[2025-09-26 22:32:11.031] [error] Traceback (most recent call last):

[2025-09-26 22:32:11.032] [error] File “C:\Users\satis\AppData\Local\Programs\ComfyUI\resources\ComfyUI\main.py”, line 147, in

[2025-09-26 22:32:11.033] [error] import execution

[2025-09-26 22:32:11.034] [error] File “C:\Users\satis\AppData\Local\Programs\ComfyUI\resources\ComfyUI\execution.py”, line 15, in

[2025-09-26 22:32:11.035] [error] import comfy.model\_management

[2025-09-26 22:32:11.036] [error] File “C:\Users\satis\AppData\Local\Programs\ComfyUI\resources\ComfyUI\comfy\model\_management.py”, line 237, in

[2025-09-26 22:32:11.037] [error] total\_vram = get\_total\_memory(get\_torch\_device()) / (1024 \* 1024)

[2025-09-26 22:32:11.055] [error] File “C:\Users\satis\AppData\Local\Programs\ComfyUI\resources\ComfyUI\comfy\model\_management.py”, line 187, in get\_torch\_device

[2025-09-26 22:32:11.056] [error] return torch.device(torch.cuda.current\_device())  
[2025-09-26 22:32:11.074] [error] File “C:\ComfyUI.venv\Lib\site-packages\torch\cuda\__init_\_.py”, line 1071, in current\_device

[2025-09-26 22:32:11.075] [error] _lazy\_init()  
[2025-09-26 22:32:11.076] [error]  
File "C:\ComfyUI.venv\Lib\site-packages\torch\cuda\_init_.py", line 403, in \_lazy\_init

[2025-09-26 22:32:11.077] [error]  
[2025-09-26 22:32:11.078] [error] raise AssertionError(“Torch not compiled with CUDA enabled”)

[2025-09-26 22:32:11.079] [error] AssertionError: Torch not compiled with CUDA enabled

---

<div class="post-metadata">

### Author: ![filtered](https://yyz1.discourse-cdn.com/flex009/user_avatar/forum.comfy.org/filtered/32/149_2.png) [@filtered](https://forum.comfy.org/u/filtered)
#### Post date: [September 28, 2025, 7:45pm UTC](https://forum.comfy.org/t/unable-to-start-comfyui-cida-ossie/3822/2 "2025-09-28T19:45:27Z")

</div>

### NVIDIA Card

If you have an NVIDIA card, please click in to **Troubleshooting** and reinstall your virtual environment. You will likely need to click the **All** filter up the top.

Cause: For some reason you have CPU-only torch installed.

* * *

### Other GPUs

If you **DO NOT** have an NVIDIA card, please ensure ensure ‘cpu’ is in your launch arguments in **ComfyUI\user\default\comfy.settings.json** (this folder is in **Documents** by default, but logs suggest you installed to **C:\ComfyUI** ):

```auto
  "Comfy.Server.LaunchArgs": {
    cpu: ""
  },

```