Browse categories
Explore
Fiverr Pro
English
$
USD
I will build custom comfyui workflows for ai image and video generation
Skai
About this gig
Struggling to get consistent, high-quality results from AI image tools? I build custom ComfyUI workflows that give you full control over every generation no more guessing, no more broken prompts.
I specialize in:
Custom text-to-image workflows (FLUX, SDXL, SD3.5)
ControlNet integration (pose, depth, Canny, IP-Adapter)
LoRA training and integration for character/brand consistency
Batch processing for e-commerce product photos
Upscaling and enhancement (SUPIR, Detail Daemon)
API-ready workflows for automated pipelines
What you get:
Complete workflow JSON file ready to load into ComfyUI
API-format JSON plug directly into your app or backend
Sample images demonstrating the workflow
Step-by-step setup guide (PDF)
Docker deployment package (Premium tier)
Why choose me:
1+ year of hands-on ComfyUI experience with 3000+ custom nodes
I deliver production-grade workflows, not just prompt presets
API integration ready connect to RunPod, Replicate, or your own GPU
Fast response and clear communication
I use the latest models: FLUX.1, SDXL, WAN 2.2, LTX-2.3, and more.
Send me a message before ordering to discuss your project. I'll recommend the best workflow setup for your needs.
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 Skai
Skai
I will build custom ComfyUI workflows for AI image and video generation
- FromChina
- Member sinceAug 2026
- Avg. response time2 hours
Languages
English
ComfyUI workflow developer specializing in production-grade AI image and video pipelines. I build custom workflows that give you full control over every generation — from FLUX text-to-image to ControlNet pose guidance, LoRA training, and batch processing.
What I work with:
• ComfyUI + 3000+ custom nodes
• FLUX.1, SDXL, SD3.5, WAN 2.2, LTX-2.3
• ControlNet, IP-Adapter, LoRA training
• Serverless API deployment (RunPod, Replicate, Docker)
I deliver workflow JSON files, API integration packages, and deployment setups — not just images.
Message me to discuss your project.
FAQ
Do I need a powerful GPU to use the workflow?
Not necessarily. You can run ComfyUI locally if you have a GPU (8GB+ VRAM recommended), or I can deploy the workflow to cloud GPUs like RunPod or Replicate so you can trigger it via API without any local hardware.
What models do you support?
I work with FLUX.1 (dev/schnell/pro), SDXL, Stable Diffusion 3.5, WAN 2.2 (video), LTX-2.3 (video), and most community models on CivitAI. If you have a specific model in mind, message me first.
Can you train a custom LoRA for my character or product?
Yes. I can train LoRA models from your image dataset (minimum 10-15 images for characters, 30+ for products). The trained LoRA file is delivered in safetensors format along with recommended strength settings and prompt templates.
Do you provide the API integration file?
Yes. The Standard and Premium tiers include an API-format JSON file that allows you to trigger the workflow programmatically via HTTP request. The Premium tier also includes Docker deployment files and documentation.
Can I request a specific use case not listed in your gig?
Absolutely. Message me with your requirements — e-commerce product photography, AI influencer generation, manga/illustration, video generation, or any custom pipeline. I'll assess feasibility and provide a custom quote.
