LoRA of Dasha Taran (For FLUX)

Description

This LoRA trained using kohya_ss on 60 images 1024×1024 up to 12000 steps (50 epochs, 240 steps each) without captions and without regularization images.
Source model: ‘Source model: flux1-dev’. This LoRA would works well with flux1-dev and flux1-schnell. Also this LoRA should works well with most of community trained FLUX family checkpoints.
learning rate: 0.00005.
Precision: fp16.
LR Scheduler: constant.

Below is example of testing AIDA_FXLR_dataran LoRA in Comfy_UI with flux1-dev:
Click on the image to view or download it in the original size.

Below is example of testing AIDA_FXLR_dataran LoRA in Comfy_UI with flux1-schnell:
Click on the image to view or download it in the original size.

Below is download link for Comfy_UI .json config to reproduce this workflow:
      2025-03-19-220628_flux1-dev_921762068039932.json_.zip
I packed it into .zip because my site can’t host .json files.

Download link to AIDA_FXLR_dataran.safetensors will be available for you after purchase. For your comfort I packed AIDA_FXLR_dataran.safetensors into .zip archive with AIDA_FXLR_dataran.png and AIDA_FXLR_dataran.txt, preview and description are compatible with AUTOMATIC1111 extra networks menu.

Please pay at least 1$ If you want to achieve rights for selling images generated with AIDA_FXLR_dataran.safetensors. Also this way you can say me special thanks for this LoRA.

Please contact me if you need some extended rights for sharing or selling AIDA_FXLR_dataran.safetensors, merging AIDA_FXLR_dataran.safetensors with other LoRA, using AIDA_FXLR_dataran.safetensors in generation services and so on.