Basic
$10/ month
100k credits ≈ 3 hours*
- Normal priority in queue
- Low concurrency
- L4 24GB included
The next gen Serverless ComfyUI Cloud
No downloads or installs are required. Zero setups.
Pay only for active GPU usage, not idle time. Zero wastage.
Open Source.
Get started for free!Comparison of billable seconds
GPU Usage
Provisioned GPUs
Pay for both active and idle GPU usage
Serverless GPUs
Pay only for active GPU usage
Serverless GPU pricing
Start exploring for free. Upgrade when you're ready.
New accounts include 5k free credits—no credit card required. Credits are spent only while a workflow is running.
Standard
$30/ month
315k credits ≈ 10 hours*
+5% bonus credits
- Higher priority in queue
- Higher concurrency
- L40S 48GB included
Pro
Most popular$60/ month
660k credits ≈ 20 hours*
+10% bonus credits
- Highest priority in queue
- Highest concurrency
- H100 80GB included
Pro Plus
$120/ month
1.4M credits ≈ 43 hours*
+15% bonus credits
- All Pro features
- Organizations
- Projects
- Analytics
Pro Max
$240/ month
2.9M credits ≈ 89 hours*
+20% bonus credits
- All Pro features
- Organizations
- Projects
- Analytics
Enterprise
Custom
Tailored infrastructure and managed operations for larger teams.
- Private clusters
- User management
- Analytics
As a fully managed platform, custom node installation is not supported. The shared executor includes 95+ extensions, 4,000+ nodes and 2,000+ models.
* Hours estimated using the L4 GPU tier.
Per-second rates
GPU pricing
Credits are spent per second of GPU execution—you're never charged for idle time.
| GPU tier | VRAM | Credits / sec | Per second | Available from |
|---|---|---|---|---|
| L4 | 24GB | 9 | $0.0009 | Basic |
| L40S | 48GB | 32 | $0.0032 | Standard |
| H100 | 80GB | 64 | $0.0064 | Pro |
10,000 credits = $1 · credits come with a monthly plan and don't roll over.
From our users
Don't take our word for it
Real testimonials from people running ComfyUI on ComfyICU.
"I burned through $50 in runpod credits messing with ComfyUI before and it just struck me as so inefficient that I was reserving these massive GPUs for hours at a time when I only needed them for like 1% of that, so thrilled to see this kind of thing gain some traction"
"before ComfyICU i tried setting up it myself via VPS / docker container ... it worked but was way way way more expansive ... I also tried setting up via serverless GPU services which was also not cheap ... in the end switching to ComfyICU was super simple"
"This platform is really helping by the way, I have spent 2 weeks trying to get a serverless implementation of this workflow running! So thanks!"
"way, way cheaper than replicate and way, way better developer experience because the ComfyUI web frontend and the API are so seamlessly integrated. Switching from API to web ui to iterate is a game changer for me."