Upwork home
Sign up
Project Catalog
DesignDesign
Video & AudioVideo & Audio
Development & ITDevelopment & IT
MarketingMarketing
Writing & TranslationWriting & Translation
MoreMore
Aviv isn't taking new orders for this project right now. Here are some similar projects to explore.

You will get ComfyUI Setup Locally or on RunPod

Aviv K.Status: Offline
Aviv K.
5.0
Top Rated
Preview image

Let a pro handle the details

Buy Generative AI services from Aviv, priced and ready to go.
Aviv K.Status: Offline
Aviv K.
5.0
Top Rated
Preview image

Let a pro handle the details

Buy Generative AI services from Aviv, priced and ready to go.

Project details

Run ComfyUI workflows locally or on cloud without the learning curve nightmare.

I'm a Top-Rated Upwork engineer (100% JSS, 8 yrs DevOps + 3 yrs Gen-AI) who has guided 150+ creators through ComfyUI's complexity maze.

What makes it different?
 • Streamlined setup: Skip the node spaghetti. Get a clean, optimized ComfyUI installation in hours, not weeks.
 • Performance-first configuration: Custom nodes, checkpoints, and LORAs selected for your specific use case and hardware limits.
 • Knowledge transfer included: Prompt engineering fundamentals, model selection strategy, and upscaling techniques that actually work.
 • Flexible deployment options: Local GPU setup or cloud setup on RunPod.

If you need ComfyUI running efficiently without months of trial-and-error, plus someone who can walk you through the basics without making it sound like technical mumbo jumbo, you're in the right spot!

Let's build workflows that work.
AI Algorithms
Convolutional Neural Network
AI Applications
AI Content Creation, AI Text-to-Image, AI-Generated Art, AI-Generated Video, Image Upscaling, Image-to-Image Translation
AI Development Language
Python
AI Tools
Hugging Face
AI Models
Stable Diffusion
What's included
Service Tiers Starter
$120
Standard
$170
Advanced
$250
Delivery Time 2 days 3 days 4 days
Number of Revisions
111
AI Model Integration
Batch Normalization
-
-
-
Database Integration
-
-
-
Detailed Code Comments
-
-
-
Image Upscaling
MLOps
-
-
-
Model Deployment
-
-
-
Model Documentation
Model Monitoring
-
-
-
Model Testing & Optimization
-
-
-
Model Tuning
Natural Language Processing
-
-
-
NLP Tokenization
-
-
-
Pre-Training
-
-
-
Prompt Engineering
Setup File
Source Code
-
-
Optional add-ons You can add these on the next page.
Fast Delivery
+$70 - $150
Additional ComfyUI Workflow (+ 1 Day)
+$70
Additional 30min 1:1 session
+$45
5.0
46 reviews
100% Complete
(46)
1% Complete
(0)
1% Complete
(0)
1% Complete
(0)
1% Complete
(0)

JE

Joseph E.
5.00
Jan 26, 2025
Rarely in my over $55k+ and hundreds of hires on Upwork have I had someone work as quickly and professionally as with my experience with Aviv. The time/value of his service is unparalleled and I am hopeful that we will have the opportunity to work again in the near future.
Aviv K.Status: Offline

About Aviv

Aviv K.Status: Offline
AI Image/Video Infrastructure & UGC Engineer | ComfyUI | RunPod
100% Job Success
5.0  (46 reviews)
Tel Aviv, Israel - 9:55 am local time
Your ComfyUI workflow runs fine locally, then breaks the moment it becomes an API on RunPod. That's most of what clients hire me for. The last backend I built served 45,000+ jobs at 99.9% uptime.

People usually reach me with one of three problems:

A workflow that runs on your machine but not behind an API or on serverless.
A product that needs a generation backend you can depend on, not another prototype.
A diffusion, LoRA, voice, or video pipeline with quality, identity, speed, or cost problems that need real debugging.

I work across Krea 2, Z Image, Qwen Image, Wan 2.2, LTX, Flux, Stable Diffusion, and Seedance 2. The model changes every few months. The hard parts don't: model loading, GPU memory, workflow graphs, reference inputs, queue workers, retries, storage, and what a single job actually costs you.

I also move fast, and clients tend to bring it up before anything else. I've delivered ahead of schedule more than once, and I answer messages in hours rather than days.

Then I explain what I built. Most of my clients aren't ComfyUI experts and have no interest in becoming one. You get a walkthrough of how the thing works and how to run it without me, and I stay reachable after delivery. A lot of my repeat work started with someone who just needed the problem explained clearly.

The creative side isn't theory for me either. I produce AI UGC and social content: influencers, consistent characters, product ads, talking-head video with voice cloning. Outside client work I train and publish open-source ComfyUI workflows and LoRA models, with 180,000+ combined downloads. That's why I can tell you why a character drifts between shots, or why footage that looks perfectly good still fails as UGC.

70+ jobs completed here, across ComfyUI workflow development, RunPod setup, serverless image and video endpoints, LoRA training, AI influencer pipelines, and custom AI media tools.

Stack: ComfyUI, RunPod, Python, FastAPI, TypeScript, Next.js, LoRA training, serverless GPU inference, workflow APIs, ElevenLabs, automation, cloud storage, diffusion and video generation systems.

Not sure whether your problem is a quick fix or a rebuild? Book a 30-minute consultation and I'll tell you straight. A lot of my longer projects started there, and a few ended there too, because the fix turned out to be small.

Steps for completing your project

After purchasing the project, send requirements so Aviv can start the project.

Delivery time starts when Aviv receives requirements from you.

Aviv works on your project following the steps below.

Revisions may occur after the delivery date.

Hardware compatability

(Ignore if you're opting for a cloud solution) We will make sure your hardware is compatible for running ComfyUI locally with your desired models.

Review the work, release payment, and leave feedback to Aviv.