This lora was trained on images from the game Granblue Fantasy.V4: Trained on 47 images. The base for this lora is <a target="_blank" rel="ugc" href="https://civarchive.com/models/2458426/anima">Anima 1.0</a>.V3: Trained on 51 images. The base for this lora is <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1190596">Noob 1.0 V-Pred</a>.V2: Trained on 80 images. The base for this lora is <a target="_blank" rel="ugc" href="https://civarchive.com/models/852957/zootlustrious-aesthetic-illustrious-xl-refinement">Zootlustrious V2.</a>V1: Trained on 160 images. The bsee for this lora is <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl">Pony</a>.To get the results you see in the images above, <a target="_blank" rel="ugc" href="https://civarchive.com/articles/11427/my-current-sdxl-settings">these are the settings I used.</a>Please note, this is a style lora. It's not mean to recreate any of the characters from the game. Pony is bad at characters as well.

Granblue Fantasy Style | Anima/SDXL

This version was trained on <a target="_blank" rel="ugc" href="https://civitai.red/models/2458426/anima">Anima 1.0</a>, at a resolution of 1792x1792. So you can gen up to and around that dimension.I had to redo the entire dataset for this lora, so pretty much none of the SDXL data was used in this lora. I had to be careful with the training since Anima picks up so much detail. 95% of the data is full body shots with white backgrounds. You can expect this lora to struggle with backgrounds to a degree.AttentionEvery once and a while, an image will pop up with strange boarders. Please add border to the negative prompt to deal with this issue.

Files with SHA256: 07adbf7b332c54605e0ff6cb7a18274adbdfabc356906a476664a4d7a360c9b8