Browse categories
Explore
Fiverr Pro
English
$
USD
I will do lora training and deploy comfyui workflow into serverless api, runpod
David C
About this gig
I will train a custom LoRA AI influencer model focused on realism, consistency, and long-term reuse using SDXL, FLUX, and ComfyUI workflows. This service is built for creators, brands, and agencies that need a stable AI identity for social media, marketing, or scalable contentwithout face drift or overfitting.
Your LoRA is trained from your image dataset using a controlled process that prioritizes data quality, correct tagging, and proper training parameters. The result is a model that preserves identity while allowing flexible styles, poses, lighting, and outfits. Youll also receive a ready-to-run ComfyUI workflow, so you can generate results immediately without rebuilding nodes or guessing settings.
Software Used
ComfyUI, Stable Diffusion XL (SDXL), FLUX, Kohya tools, Python training pipeline.
Deliverables
Custom LoRA (.safetensors)
ComfyUI workflow (.json)
Prompt notes (.txt)
Test images (.png / .jpg)
I focus on clean training and usable workflowsnot presets.
Message me before ordering to review your dataset and choose the right package.
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 David C
David C
ComfyUI workflows, automation, and custom AI models
- FromUnited States
- Member sinceFeb 2026
- Avg. response time4 hours
- Last delivery2 months
Languages
English, Spanish, German, French
I specialize in building custom ComfyUI workflows, AI image automation, and tailored LoRA model training for Stable Diffusion, Flux, and Wan 2.0. My goal is simple: help you achieve consistent, reliable AI image results without confusion, broken setups, or trial and error. I create clean, production-ready workflows with the latest ComfyUI updates, including optimized node structures, batching, scheduler tuning, and performance enhancements—delivering practical, reusable systems that are easy to load, scale, and use.
Other AI Artists Services I Offer
FAQ
What do you need from me to start?
I need a clear image dataset (15–40 images), reference details, and your intended use case. Contact me before ordering to confirm suitability.
Can you train from low-quality or AI-generated images?
No. For best results, images must be clear, consistent, and preferably real photos. Poor datasets reduce model accuracy.
Which platforms and models do you support?
I train LoRA models for SDXL and FLUX using ComfyUI-based workflows only.
Will I receive the workflow files?
Yes. You receive the trained LoRA and a ready-to-run ComfyUI workflow.
Can I request revisions?
Revisions are limited to training adjustments only. Dataset changes require a new order
