Training a Lora on A1111? Is it possible? #15321
-
|
Hi folks, is it possible to train a lora on A1111? |
Beta Was this translation helpful? Give feedback.
Answered by
dilectiogames
Mar 19, 2024
Replies: 1 comment 1 reply
-
|
theoretically yes, you can use the dreambooth extension. In practice I've never managed to train anything on automatic1111 because the CUDA and the Vram are not enough because they are reserved by automatic. Instead use kohya_ss (is a separate installation) which works fine with the same resources that automatic + dreambooth said it was not enough |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
Thanks mate! |
Beta Was this translation helpful? Give feedback.
Answer selected by
suonnon
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
theoretically yes, you can use the dreambooth extension. In practice I've never managed to train anything on automatic1111 because the CUDA and the Vram are not enough because they are reserved by automatic.
Instead use kohya_ss (is a separate installation) which works fine with the same resources that automatic + dreambooth said it was not enough