Browse categories
Explore
Fiverr Pro
English
$
USD
I will train a custom anima, illustrious lora for anime and video game characters
chitra
About this gig
I train custom LoRA models for anime, video game characters, and Original Characters (OCs) using base models like Anima and Illustrious XL.
Technical Process
- Dataset Engineering: Raw frame extraction and AI upscaling for low-res images.
- Preprocessing: Precise cropping, aspect ratio correction, and manual editing via Krita to remove noise.
- Tagging: Danbooru tagging with isolated trigger words to keep facial features, hair, and outfits modular without concept bleeding.
- Model Tuning: Network dimensions and step calculations adjusted per concept complexity.
Service Scope
- Single Concept: 1 character, object, or outfit.
- Multi-Outfit: Up to 3 outfits per character. Max 100 images.
- Dynamic Concepts: Complex geometries such as action poses, object interactions, or mechanical transformations.
Deliverables
- Trained .safetensors LoRA file.
- Generation prompts and trigger word lists.
- Sample images showing model fidelity.
Contact me before ordering if your dataset needs building from scratch.
Delivery style preference
Please inform the freelancer of any preferences or concerns regarding the use of AI tools in the completion and/or delivery of your order.
Get to know chitra
chitra
- FromIndia
- Member sinceJun 2024
Languages
Hindi, English
I train custom LoRA models using base models Illustrious XL and Anima.
What I Do
Character LoRAs: Experienced in training anime and video game characters with high accuracy.
Complex Poses: Can build LoRAs for specific action poses, character interactions, and dynamic angles.
Dataset Quality: Complete manual dataset cleanup, precise Krita editing, and strict Danbooru tagging to prevent concept bleeding.
