LoRA in the style of Rui Araizumi, legendary illustrator of the Slayers and Lost Universe series. Mixes both her original retro style and the new "modern retro-emulating" style.
Due to dataset availability this LoRA is like 90% Lina Inverse at this point. I'll need some additional sources to find some guys lol. Still produces various outputs well, just overfit. Next version.
Find an index of all my LoRAs here.
If you like the style, support the artist Rui Araizumi, or the Slayers/Lost Universe anime/games. Supporting the original artists is crucial.
Please use this model only for casual everyday use cases, like mocking up drawings. Please don't use it in a corporate/professional way that competes with the original author or hurts their career/business. Use AI responsibly.
I do not profit from making any model, do not sell any image or model, and do not take commissions of any sort. The model permissions are also set so that you cannot sell images or merges of this model. Again, use AI responsibly.
(Model will, naturally, be removed immediately if requested by author/entity whose style is being replicated)
Description
Timestamp : 20260807_1631
Dataset : ruiaraizumi_rzms
Dataset Path : /workspace/datasets/ruiaraizumi_rzms
Images : 286
Captions : 286
Batch Size : 2
Target Steps : 1001
Steps / Epoch : 143
Epochs : 7
Rank : 96
Alpha : 64
Learning Rate : 0.0001
Trigger Token : rzms
Base Model : /workspace/models/Anima-Base-v1.0-Diffusers
Workspace : /workspace/TrainingRuns/TrainingRun_ruiaraizumi_rzms_20260807_1631
Cache : /workspace/cache/ruiaraizumi_rzms
Output Model : /workspace/TrainingRuns/TrainingRun_ruiaraizumi_rzms_20260807_1631/Anima_Base_ruiaraizumi_rzms.safetensors
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.





