FLUX.1-dev lora compatibility
Thank you very much for the FLUX.1-Kontext-dev release! Can you comment on whether it is possible/reasonable to use FLUX.1-dev loras with the Kontext model? I.e., is the architecture compatible in terms of lora usage, at least in principle, and also in practice? What is the relation between the two models in this respect? Many thanks in advance for any clarification and additional information you can provide!
I'm not sure if it will be useful to you, but I checked the generation using two Lora: Flux.1-Turbo-Alpha, Cyberpunk Anime Style.
Quality of the result is pretty fine (almost), and the generation speed has increased (1.5 min, instead of 3+), and the image definitely has a cyberpunk style. So in a way it's compatible and it works. But my test sample is not large yet.