LoRA fine-tuning of a 12B-parameter image generation model (Krea 2 / Qwen-Image MMDiT) for a personally-created character, on a single consumer-grade 12 GB GPU. This repository accompanies the ...
parameters are created (or updated if they already exist) in the destination part. Only user-accessible parameters (Read/Write) are shown. Formulas are not copied — only the current numeric or string ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results