Browse categories
Explore
Fiverr Pro
English
$
USD
I will create a consistent ai anime character with a custom lora
Anthony Aora
About this gig
IMPORTANT: PLEASE READ BEFORE ORDERING
I create custom anime character LoRAs using my AoRA pipeline morphology-aware training focused on appearance and face consistency across generations.
Basic: You provide a complete, ready-to-train dataset. I handle training and delivery.
Standard & Premium: You provide 15 official reference images. My pipeline automatically harvests and curates the rest of the dataset.
Please contact me before ordering for: rare characters, original characters, complex styles, or if you are unsure about your images. I review your material before confirming.
LoRA quality depends on the clarity and consistency of your reference images especially for the Basic package.
Revisions cover prompt and weight adjustments only. A new character, new dataset, or full restart is not included unless agreed separately.
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 Anthony Aora
Anthony Aora
Premium custom LoRa Training High End AI Image Generation
- FromFrance
- Member sinceMay 2026
Languages
English, French
Welcome! I’m an AI Content Specialist focused on creating high-quality custom LoRAs (SDXL/Anima) and detailed AI-generated illustrations.
What sets me apart? I go beyond simple image generation. I use advanced workflows to analyze and refine character anatomy, including proportions, ratios, and unique body types. This allows me to produce accurate, consistent, and visually appealing results—without the common distortions seen in standard AI outputs.
FAQ
What is AoRA?
AoRA is my custom training approach focused on character morphology — appearance proportions, face accuracy and pose consistency across all generated images.
How many images do I need? (BASIC ONLY)
for basic only : Provide 30–50 clean reference images original and fanart. Prioritize neutral poses (A-pose or standing), minimal clothing or form-fitting outfits. Avoid motion blur, heavy stylization, or baggy clothing that hides appearance proportions. Image quality directly impacts LoRA quality.
Do I need to provide images?
Basic: yes, full dataset required. Standard & Premium: just 15 official reference images — my pipeline handles the rest. Unsure? Message me before ordering.
Can you train original characters?
Yes, but please contact me before ordering. Original characters need clear references and consistent visual details.
Is the LoRA file included?
Yes. Every package includes a .safetensors file, ready to use in A1111, ComfyUI, or Forge.
Can I use the LoRA commercially?
The LoRA is for personal use only. For commercial use, contact me before ordering so we can discuss terms.
Do revisions include retraining from zero?
No. Revisions cover prompt and weight adjustments only. A full retrain counts as a new order.
