This image generator uses Illustrious model as its base Checkpoint, combines a few LoRAs and some weighted quality tags to produce manga‑like images. The most notable parameters, including FaceDetailer and HiresFix, significantly improve the quality of the output image. Input Suggestions: Since Illustrious model is an SDXL variant and a booru‑based model, inputs are best written using booru taglines, e.g. "1girl, solo, black hair, medium hair, black eyes, ...". A booru‑based model also understands properly written English sentences, but to a lesser degree.
