You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have trained a flux lora with simpletuner using dim=32, alpha=1. I trained with dim=32 and alpha=32 and it works OK, but with alpha=1, it generate noise. But in the simpletuner, when evaluate, it uses flux pipeline and it can generate good image.
I have upgraded the latest version of the comfyui.
Actual Behavior
This is the image generated in the simpletuner with flux pipeline.
Steps to Reproduce
This image the image generated in the comfyui, just noise
Debug Logs
there is no bug information
Other
No response
The text was updated successfully, but these errors were encountered:
Expected Behavior
I have trained a flux lora with simpletuner using dim=32, alpha=1. I trained with dim=32 and alpha=32 and it works OK, but with alpha=1, it generate noise. But in the simpletuner, when evaluate, it uses flux pipeline and it can generate good image.
I have upgraded the latest version of the comfyui.
Actual Behavior
This is the image generated in the simpletuner with flux pipeline.
Steps to Reproduce
This image the image generated in the comfyui, just noise
Debug Logs
Other
No response
The text was updated successfully, but these errors were encountered: