This lora was trained on about 60 images from the artist Nagi Itsuki. The images used were mostly backgrounds with a few characters in between.Anima: Captures the style pretty well. The character art is replicated pretty accurately. The backgrounds are a hit or miss. The source materials have so much small details that I had to train the lora more stronger than I normally would. Because of this, I recommend lowering the strength if the background becomes too chaotic.SDXL: The lora captures the style and shading but fails miserably to replicate the details of the source images backgrounds. You will notice this if you attempt to prompt a large scenery. I feel this is a limitation of SDXL more than anything.I recommend you stick to close-up and simple concepts. The character style replication works well enough but it becomes influenced by the model once you begin to prompt concepts outside of the artist's territory. The actual character art of the source is much more simple than what you'll get from this lora due to the model. If you want something more simple similar to the artist's character style, I recommend using less score tags for pony.Version 3.0 was trained on <a target="_blank" rel="ugc" href="https://huggingface.co/circlestone-labs/Anima">Anima V1.0</a>.Version 2.0 was trained on <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1190596">Noob 1.0 V-Pred</a>.Version 1.0 was trained on <a target="_blank" rel="ugc" href="https://civarchive.com/models/288584?modelVersionId=324619">AutismMix</a> and I haven't tested on anything else.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>Visit My Site For Models I Can't Post On CivitAI: <a target="_blank" rel="ugc" href="https://qtgurl.com/">https://qtgurl.com/</a>

Files with SHA256: bb6f7f51d81fde7d13d9157dd4fc1ae995658615107bf6f121d8e2b994744111