Train FLUX.1-Fill-dev LoRa
Hi, I want to train a LoRa and I found this fork of the diffusers git https://github.com/Sebastian-Zok/FLUX-Fill-LoRa-Training/tree/main?tab=readme-ov-file . Has anyone successfully trained a LoRa with this model? If so, what hardware did you use and what were the settings you were running?
Hi, I want to train a LoRa and I found this fork of the diffusers git https://github.com/Sebastian-Zok/FLUX-Fill-LoRa-Training/tree/main?tab=readme-ov-file . Has anyone successfully trained a LoRa with this model? If so, what hardware did you use and what were the settings you were running?
In one trainer pick the flux lora training and then choose the flux dev fill model.
4080 works just fine with batch size 1
Thank you very much!
does this actually work and give as good results as dev fill normally does for inpanting?