Browse categories
Explore
Fiverr Pro
English
$
USD
I will set up comfyui ai image and video models on runpod or your GPU
Jubied
About this gig
Need local AI image or video generation running as a tested ComfyUI solution, not just an installation?
I set up and validate local or open-weight models on RunPod, NVIDIA GPU cloud, a Windows or Linux GPU PC, or compatible Apple Silicon when the workflow supports it. Stacks can include MiniMax H3, Wan, LTX Video, Krea 2, FLUX Klein, Qwen, GGUF or quantized models, and custom nodes.
EVERY PACKAGE INCLUDES
- Hardware, VRAM, license, and feasibility check
- Scoped environment, dependencies, and model setup
- One agreed workflow or model path
- Live test with representative input
- Sample output, model and node list, and startup instructions
I can use Tavris1 ComfyUI Easy Install where appropriate, or a manual or container setup when it is more reliable. Standard adds a persistent environment and optimization. Premium can add a scoped API, serverless flow, storage, queue, logging, or webhook when listed.
Eight GB VRAM is only an entry point for selected light or quantized workflows; it is not enough for every model. GPU, storage, API, and licensed-model costs are buyer-paid. Speed, quality, uptime, and compatibility are confirmed only through the agreed test.
Get to know Jubied
Jubied
ComfyUI Workflow Engineer for Image, Video and Automation
- FromBangladesh
- Member sinceMar 2020
- Avg. response time2 hours
- Last delivery1 month
Languages
English, Arabic, Portuguese, Spanish, German, French, Macedonian
I'm Jubied, founder of ComfyHelper. I build custom ComfyUI image and video workflows for creators, agencies, and brands. My work covers Flux/SDXL, LoRAs, product photography, character pipelines, and batch automation, with API integrations and cloud GPU deployment.
Each project starts with a feasibility check and written scope. You receive the agreed workflow files, sample outputs, dependency list, and setup documentation, tested for the agreed environment.
Send your use case, sample inputs, target output, GPU/OS, and deadline. I will confirm scope, timeline, and price.
My Portfolio
FAQ
Which AI models can you install?
MiniMax H3, Wan, LTX Video, Krea 2, FLUX Klein, Qwen, GGUF or quantized models, and other compatible stacks after hardware, license, and node checks.
Where can you deploy the solution?
RunPod, an agreed NVIDIA GPU cloud, a Windows or Linux dedicated-GPU PC, or compatible Apple Silicon for supported workflows.
Is 8 GB VRAM enough?
Only for selected light or quantized workflows. Large video models often need much more. I check the exact model, resolution, batch size, and GPU first.
Do you use ComfyUI Easy Install?
I can use the Tavris1 ComfyUI Easy Install Windows or NVIDIA route, or its MAC-Linux branch, when appropriate. I use a manual or container path when more reliable.
What will I receive?
A configured environment, agreed workflow or model path, dependency and model list, representative live test, sample output, and startup instructions. Higher packages add the listed production elements.
Are GPU, storage, API, or model costs included?
No. The buyer pays cloud GPU time, storage, bandwidth, API credits, and any licensed model or weight costs.
How should I provide server or cloud access?
Use temporary, least-privilege access through an approved secure method. Never send your password. Revoke or rotate access after handoff.
What does a revision cover?
A revision fixes or adjusts the agreed setup. A new model, workflow, provider, API, deployment target, or integration is new scope.
Does this package include an autonomous AI agent?
No. Fiverr's package field represents the single AI solution being implemented. An autonomous agent, chatbot, or LLM orchestration layer is included only when written in a custom offer.
