This is the second Flux LoRA that I’ve trained, and this time I chose Schnell. I experimented with combining two different types of captions for training: one being natural language and the other being tags. Therefore, when using prompts, you might get better results by combining both keywords and natural language. After this round, I’ve gained some valuable experience. Next time, I’ll train Rebecca using a different method and some other parameters to see which approach is ultimately the most optimal in terms of time, cost, and quality. you like my content, support me at:<a target="_blank" rel="ugc" href="https://ko-fi.com/duchaiten">https://ko-fi.com/duchaiten</a>Patreon in here: <a target="_blank" rel="ugc" href="https://linktr.ee/Duc_Haiten">https://linktr.ee/Duc_Haiten</a>If you want to use my models to create unlimited images at a small cost without having to install anything, <a target="_blank" rel="ugc" href="http://mage.space">mage.space</a> will be a good choice, i have some models that are exclusively available on there.

Files with SHA256: 7de68d4c4dfa14c4cbca878543ae716f02b93976a25ee52060fb5ac96f1d8399