Kontext doesn't work well with lora when rank is large
- Dominant language
- Python
- Stars
- 133k
- Forks
- 15.7k
- Avg merge
- 1d 7h
- Merged PRs (30d)
- 158
Description
### Custom Node Testing
- [x] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)
### Expected Behavior
Kontext should work well with the lora loader, whatever the lora rank is
### Actual Behavior
When the lora rank is 4 or 8, the results look great. However, when the lora rank is 16 or 32, the results become to noise. I also tried the same lora with diffusers, and everything was fine. Btw, if I used diffusers to merge lora into the transformer and then loaded this model in comfyui, the results would also be correct.
### Steps to Reproduce
None
### Debug Logs
```powershell
None
```
### Other
_No response_
Contributor guide
Assessment
This issue has not been assessed yet.