Choose the perfect plan for you

Hobby

No Subscription
Standard machine rates apply

Features you'll love:

  • Free trial credits included
  • Dedicated and Private 16-80GB VRAM GPU Servers
  • Native ComfyUI in sync with official releases
  • Upload custom models, install nodes freely
  • Save workflows and their environments for later use and sharing
  • Deploy workflows as serverless APIs instantly
  • Concurrent multi-workflow runs on GPUs
  • 10GB private storage, 72-hour persistence
  • 200+ ready-to-run workflows for prototyping
Yearly
-33% off
Monthly

Pro

$29.99$19.99
billed $239.9/year (4 months free)
Over 20% discounted machine rates

Everything in Hobby, plus:

  • $10 balance credit every month
  • Over 20% discount on all machine rates, covering both ComfyUI cloud usage and serverless API inference calls for cost-effective scaling
  • 20 FREE CPU hours monthly for file/workflow setup
  • 200GB private storage, lifetime persistence
  • Priority Discord and email support

Compare Plan Features

Machine RatesHobbyPro
Small Machine: 20 hours/month free CPU time
For file upload/download and workflow setup
Free
Medium: T4, A4000
16GB VRAM | 16GB RAM | 8 vCPUs
$0.99/hr
$0.79/hr
Large: A10G, A5000
24GB VRAM | 32GB RAM | 8 vCPUs
$1.75/hr
$1.39/hr
X-Large: A6000
48GB VRAM | 48GB RAM | 28 vCPUs
$2.5/hr
$1.99/hr
X-Large Plus (API ONLY): L40S, L40
48GB VRAM | 64GB RAM | 28 vCPUs
$2.99/hr
$2.15/hr
2X-Large: A100
80GB VRAM | 96GB RAM | 28 vCPUs
$4.99/hr
$3.99/hr
2X-Large Plus: H100
80GB VRAM | 180GB RAM | 28 vCPUs
$7.49/hr
$5.99/hr
100% private workspace
Native ComfyUI in sync with official releases
Custom models/LoRAs upload and nodes installation through Manager
Save workflows and their environments for later use and sharing
Serverless API deployment
200+ ready-to-run workflows
User storage capacity10GB200GB
Persistent storageFiles deleted after 72 hrs inactivity
Monthly balance credit
SupportDiscord/Email supportPriority support

See why artists like you choose RunComfy

Frequently Asked Questions

Is the RunComfy machine and API private?

Yes, your ComfyUI runs on a dedicated, private machine for each session, ensuring no one else can access it. All sessions and API deployments use isolated instances, keeping your data, models, and outputs exclusively yours.

Can I upload my own models/LoRAs?

Yes, you can upload your models/LoRAs securely. All uploaded and generated files are stored in private storage that is accessible only to you. This storage can be used across sessions, allowing you to use your files conveniently without the need to re-upload them each time

Can I install any custom nodes?

Yes, you can easily install custom nodes using the ComfyUI manager. While RunComfy preloads some popular extensions for your convenience, you have the flexibility to install additional nodes yourself.

Can I deploy any ComfyUI workflow as an API?

Yes, seamlessly convert any ComfyUI workflow into a serverless API for scalable, on-demand inference. It handles provisioning compute, executing the workflow, and tearing it down, all initiated via a simple API request. Perfect for easy integration with applications, scripts, or backend services. For details, check the RunComfy API Docs.

Why should I use RunComfy for my AI class/workshop/tutorials?

RunComfy is a ComfyUI Cloud service that supports AI educators by providing a seamless, collaborative, and high-performance environment for teaching ComfyUI in AI art. It offers unique features like cloud saving, extensive GPU options, flexible payment systems, and an up-to-date ComfyUI environment, making it ideal for classes, workshops, and tutorials. For more information, please visit here.

How to get invoice?

More questions?

RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Playground, enabling artists to harness the latest AI tools to create incredible art.