Browse categories
Explore
Fiverr Pro
English
$
USD
I will train a custom flux lora for your ai character
Jerry K
About this gig
Need the same AI character to look like themselves in every image? I train a custom Flux
LoRA on your images so your character stays consistent: same face, same build, same
signature outfit.
What you get: - A Flux LoRA (.safetensors) trained on flux1-krea-dev - Test renders so you
can see it works before you use it - A skin-realism pass on the renders to cut the "plastic"
AI look - A ready-to-load ComfyUI workflow and simple setup notes -
Fast delivery, starting at 24 hours
Great for original characters, mascots, AI influencers, comics, and storytelling.
How it works: send 10-30 images of your character, I build and caption the dataset, train,
score the face match, and send the files plus renders.
Please read before ordering: - This is AI-generated work that I customize for you. - You must
own the images, or have written consent from the person shown. - No celebrities, public
figures, or minors. SFW only. - No deepfakes or impersonation. I decline those orders.
Not sure which package fits? Message me first and I'll help.
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.
AI Avatar Design disclosure
I understand that by acquiring AI Avatar Design services, I must abide by any and all disclaimer rules required by applicable laws and the platform(s) where this content will be posted. Read more about Fiverr’s AI Guidelines.
Get to know Jerry K
Jerry K
Computer Engineer AI Expert Musician Producer
- FromUnited States
- Member sinceOct 2026
- Avg. response time1 hour
Languages
English
I am a computer engineer and technology executive with over 25 years of experience leading IT infrastructure, software development, and digital strategy. I specialize in generative AI, specifically training custom Flux LoRA models and building production pipelines for consistent AI characters. My background includes serving as a CTO and founding multiple technology ventures where I architected scalable digital ecosystems.
My Portfolio
FAQ
Is this AI work?
Yes. I use AI tools (Flux, ComfyUI, Kohya) and customize every order: I clean and caption your dataset, tune the training, check the face match, and run a realism pass. I'm upfront about the AI used.
Can I send photos of a real person?
Only if you own the photos and the person is an adult who gave written consent (or it's you). No celebrities, public figures, or minors, and no deepfakes. I'll cancel orders that break this.
How many images do I need?
10-30 is ideal. Clear, varied images (different angles, expressions, lighting) give better results than many similar ones. Not enough images? Message me and we'll work out a plan.
What do I receive?
A Flux LoRA (.safetensors), a trigger word and prompt guide, test renders, a ComfyUI workflow (.json), and setup notes. Standard and Premium add the realism pass and outfit consistency.
Will it work in my setup?
It's trained for Flux (flux1-krea-dev) and works in ComfyUI. It usually works in other Flux- compatible tools too works best if you have 4090 or higher NVIDA graphics card. Tell me your setup in the requirements and I'll add notes for it.
What if the character doesn't look right?
They're used only for your order. I don't share them or use them in my portfolio, and I delete them within 14 days after the order is complete unless you ask me to keep them for future revisions.
Do you make NSFW content?
No. SFW only, for training and for renders send a message if you think it borderline and I will review it,
