Verified: 7 months ago
SafeTensor
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.
An Anime-Character F.1 D LoRA Model
The character is Ayase Eli who comes from the series called "Love Live! School Idol Project".
The solution for LoRA training
Well, I just use the sd3 branch of the repo sd-scripts to train it and I tried to write the dataset config file and the parameters of training unet scripts MANUALLY. However, I couldn't find a way to obvoiusly reduce the average loss and due to the BAD performance of my laptop, I may could not upload other characters LoRA during a short time.
How to generate images like the preview images
OK, fine, I am not good at choosing the best generating parameters. I just tried the Weight of LoRA 1.25, and I'm not sure what the lowest weight value can generate a good image enough. You can strictly follow my workflow nodes and the prompts showed in the preview images.
I recommend using the words in the DeepBooru dictionary. If u want the Florence 2 version, please tell me.
这是我个人第一次尝试上传自己训练好的F.1 D LoRA，训练参数上受到fluxgym写的scripts的启发，决定自己手动调用sd-scripts的sd3分支来完成LoRA训练。
至于使用的关键词，我推荐使用来自DeepBooru的词汇，因为目前暂时没有试着用Florence 2辅助打标 （太麻烦）。
触发词很简单，就是一个ayase eli就完事，但是为了保证角色特征一致性，建议还是稍微写一下眼睛和头发的基本特征（也许也可以不用？）：
<lora:ayase-eli:1.25>,ayase eli,1girl, solo, blue eyes, long hair, blonde hair,考虑到本人的笔电性能较差，所以可能不会短时间内大批量训练很多的角色LoRA，见谅！
至于本人使用的workflow，我这里为了生成速度和更低的硬件资源占用，花了大约一整个晚上的时间，安装好了svdquant这个很强大的工具，如果你想用stable-diffusion-webui之类的，或者用ComfyUI的传统workflow，只要能兼容使用FLUX.1 Dev这个Base Model，就可以用这个LoRA！（确信