This is my Flux LoRA, and honestly, I don't have much experience in training LoRA models. You could say this was an experiment, and it truly cost me quite a bit of time and money compared to training a typical LoRA. I would say this is both a success and a failure. I was too ambitious; instead of focusing on a specific topic and trying to make it converge quickly, I attempted to turn it into a versatile NSFW LoRA, which was my mistake. This made the training process unnecessarily long for a first experiment. With a dataset of 1,000 images, it was too much for a single LoRA, but too little to improve all the NSFW concepts related to Flux. Perhaps I'll have to wait until finetuning the Flux model becomes more proficient. At that point, I'll try to fully train the FLUX checkpoint. But for now, LoRAs can only be seen as a temporary substitute. When the time comes, I might need to seek community support for a project to supplement NSFW concepts for FLUX.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.Hide

Files with SHA256: 5bb4c618bb697255921ea710e9b61fa1679714fb0e2b6554b7f5a0fb7f3b9b87