I will custom comfyui workflow ai influencer model flux lora training stable diffusion


About this gig
Need a custom ComfyUI workflow or FLUX LoRA that keeps your AI influencer or virtual model consistent across different poses, outfits and scenes?
I build reusable image-generation pipelines and train custom LoRA models using your approved reference images
WHAT I OFFER:
- Custom ComfyUI workflow setup, fixes and workflow JSON
- FLUX / FLUX.1 and SDXL LoRA training (.safetensors)
- AI influencer, virtual model and consistent character workflows
- Stable Diffusion text-to-image and image-to-image pipelines
- Dataset cleaning, image captioning and training optimization
- ControlNet, OpenPose, IPAdapter, PuLID and InstantID
- Inpainting, face detailing, upscaling and photorealistic portraits
- RunPod or local GPU setup, custom nodes and VRAM optimization
- Batch generation, prompt presets and reproducible settings
- Optional Wan image-to-video workflows and API integration
You receive tested sample outputs, the agreed workflow/model files and setup notes. I focus on reliable, editable workflows rather than one-off AI pictures
Message me with your goal, reference images and GPU setup before ordering.
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 Donald
ComfyUI and FLUX LoRA, n8n AI Video Automation
- FromUnited Kingdom
- Member sinceApr 2026
- Avg. response time1 hour
Languages
English, French, Spanish, German
My Portfolio
FAQ
Can you create a custom ComfyUI workflow from scratch?
Yes. I can build ComfyUI workflows for AI image generation, consistent characters, virtual models, portraits, product images, and supported image-to-video pipelines. Share your goal and references first.
Can you train a custom FLUX or SDXL LoRA model?
Yes. I can prepare an approved dataset, configure FLUX or SDXL LoRA training, test checkpoints and deliver the trained .safetensors file with recommended settings.
Can you create a consistent AI influencer or virtual model?
Yes. I can develop an original virtual character workflow using LoRA training, reference conditioning and consistency tests across different clothing, poses, lighting and backgrounds. Exact consistency cannot be guaranteed.
Do you work with ControlNet, IPAdapter and Stable Diffusion?
Yes. Depending on the project, I can configure Stable Diffusion, ControlNet, OpenPose, IPAdapter, inpainting, PuLID, InstantID, face detailing and upscaling to improve control over generated images.
Can you fix broken ComfyUI workflows or install them on RunPod?
Yes. I can troubleshoot missing custom nodes, model-loading errors, incompatible checkpoints, incorrect paths and GPU memory issues. I can also configure supported ComfyUI workflows locally or on RunPod.
Can you build ComfyUI workflows for AI video generation?
Yes. I can discuss image-to-video or text-to-video workflows using compatible Wan or other supported models. Video generation, motion transfer and lip-sync are scoped separately because requirements vary.
What files will I receive after delivery?
Depending on your package, you will receive an editable ComfyUI workflow JSON, custom LoRA weights, sample outputs, recommended generation settings, dependency information and setup instructions.

