Welcome to Landscape Anime Max SDXL!
LATEST VERSION: 1.1 ESTELLA+
FUTURE SUB-VERSIONS WILL COME OUT AS LoRAs INSTEAD OF CHECKPOINTS. Eg. 1.0 2.0 3.0 Checkpoints and 1.1 1.2 are LoRAs that could be used upon the version checkpoint!
This model is a derivative of the original "Landscape Anime Pro," enhanced with features from several other models, including animeIllustDiffusion v0.8 and YamersRealistic v5 and our new model's distillation outputs: EstellaXL.
Try to use the sampling method of DPM2++ a Karras or Heun at 20~50 steps for best outcomes. Clip Skip 2.
To stay updated and join our community, feel free to join our Discord server by the United DeviantArt AIGC Art Alliance: [https://discord.gg/2PfHqzxVAv](https://discord.gg/2PfHqzxVAv). Get the latest updates, or if you’d like to hear me chat in the admin channel (which is muted, but not locked 😉), come say hi!
Please note:
Landscape Anime Max SDXL is designed for image generation starting from 768x768 resolution up to 1536x1536. If you're looking to generate images at 512x512 or below, please check out the non-SDXL version of the model, Landscape Anime Pro, which is still being updated.
DISCLAIMER:
The early versions of this model may produce unstable outputs, and the quality can vary. For a more stable experience, please use Landscape Anime Pro.
Description
We have trained this version based on the Landscape Realistic Max 2.0 ≥w≤ and our new EstellaXL model as the basis.
But this version could be applied to that model or the basic SDXL model!
Illustrious is not yet available, but it is acceptable. If you want to use Illustrious please wait for the Distilled version.
So then in this version, you could also generate images with humans inside!--But only to the upper half body. We are also fixing the hand issue which we did not good in EstellaXL rather...
So please do not expect super good full body performance but we still have good image generations!