Cuda in RTX 2060
- Dominant language
- Python
- Stars
- 133k
- Forks
- 15.7k
- Avg merge
- 1d 7h
- Merged PRs (30d)
- 158
Description
### Expected Behavior
Not work Cuda on RTX 2060, maybe issue with Pytorch ver
### Actual Behavior
I have been looking for a decision for a long time, and now I decided to write here. What is the problem? Cuda nuclei are not found on my RTX 2060, they definitely eat, about 1920. I found a possible problem in the Pytorch version, but I could not find how to change it. I need it to AnimateDiff Sampler (attached json workflow)
### Steps to Reproduce
[img2vid.json](https://github.com/user-attachments/files/20259194/img2vid.json)
### Debug Logs
```powershell
D:\AIMG>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --gpu-only
[START] Security scan
[DONE] Security scan
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2025-05-17 00:00:42.376
** Platform: Windows
** Python version: 3.12.9 (tags/v3.12.9:fdb8142, Feb 4 2025, 15:27:58) [MSC v.1942 64 bit (AMD64)]
** Python executable: D:\AIMG\python_embeded\python.exe
** ComfyUI Path: D:\AIMG\ComfyUI
** ComfyUI Base Folder Path: D:\AIMG\ComfyUI
** User directory: D:\AIMG\ComfyUI\user
** ComfyUI-Manager config path: D:\AIMG\ComfyUI\user\default\ComfyUI-Manager\config.ini
** Log path: D:\AIMG\ComfyUI\user\comfyui.log
Prestartup times for custom nodes:
6.1 seconds: D:\AIMG\ComfyUI\custom_nodes\comfyui-manager
Checkpoint files will always be loaded safely.
Total VRAM 6144 MB, total RAM 32660 MB
pytorch version: 2.6.0+cu126
Set vram state to: HIGH_VRAM
Device: cuda:0 NVIDIA GeForce RTX 2060 : cudaMallocAsync
Using pytorch attention
ComfyUI version: 0.3.27
ComfyUI frontend version: 1.14.5
```
### Other
_No response_
Contributor guide
Assessment
This issue has not been assessed yet.