I will train a custom anima, illustrious lora for anime and video game characters

I train custom LoRA models for anime, game, and original characters using Anima and Illustrious XL base models. I focus on high-precision dataset curation and reconstruction to deliver maximum canon fidelity without model drift.

Core Workflow (Included in All Orders):

• Gemini Enhancement: Low-res pics are enhanced via Gemini to restore sharp facial features, clean eyelashes, and precise lineart without pixel degradation.
• Background Removal: Backgrounds and unnecessary things are cleanly removed using Gemini, so character generates seamlessly across any custom environment.
• Tag Audit: Unique outfit triggers and conflicting tags are manually verified and curated per image. The entire dataset then undergoes a final Gemini deep-scan to eliminate spelling errors, syntax bugs, and duplicates, ensuring zero concept bleed.

Premium Exclusives:

• Complete Dataset from Scratch: Just share the episode or source links. I manually select the best frames and reconstruct canon-accurate details (outfit patterns, jewelry, accessories, eyes, fingers).

Deliverables:

• Trained .safetensors LoRA file
• Complete curated dataset (images + verified .txt tags)
• Prompt guide & trigger words list