Now, people with systems that are unsupported or do not have local GPUs (like Macs or low end PCs) can run ComfyUI with ease. ComfyUI Cloud is the official cloud platform released by ComfyUI for users who lack powerful system resources such as VRAM, CPU, or SSDs.
You just need to log in using your email and password, no heavy model downloads, no long waits, and no more missing node errors from native setups. All the features are the same as those in the Desktop variant or the Portable variant.
ComfyUI Cloud comes with pre-installed models and native workflows, so you don't have to waste your valuable time loading or connecting messy workflows.
Table of Contents:
Features
1. Powerful GPU- Runs on NVIDIA A100 (40GB VRAM), great for heavy AI and image generation work. Each workflow can run for up to 30 minutes that means if it goes over, it will stop automatically.
2. Usage & Charges- You only pay for the time your GPU is actually running. You get 8 free GPU hours every day. You can buy extra hours anytime, and any unused ones carry over to the next month.
3. Models & Custom Nodes- Comes with 200+ pre-installed models ready to use. Supports the top 10 most popular ComfyUI custom nodes (more being added soon). With Partner Nodes, you can also run special or private models using the same credits. Each account includes $10 in free API credits every month, which refill automatically.
4. Privacy & Security- Your data is fully encrypted and only you can see it, even the Comfy team can't access it as reported officially.
5. Workflow Management- You can queue up to 4 jobs at once. Running multiple jobs at the same time will be added soon. Some other third party features will be added like uploading and using your own custom models, parallel runs for faster workflow execution.
Steps to join Public Beta
We have received early access to the ComfyUI Cloud Private Beta, and we are excited to test one of the most powerful workflows to see just how much potential this platform holds. Earlier, they were running the private beta that has been ended. Right now, ComfyUI Cloud is in its Public beta stage, which means its open for all and it comes with its costs.
Now whenever it clearly moves out of the beta, the team will likely introduced their pricing system. Currently, you can sign up by accessing their platform and enjoy readily official related workflows. If you are not an early bird and came late then you can do DM on reddit/X to Crystal Alpine.
We just keep on testing the Wan 2.2 Animate native setup that is one of the complex workflows. Right now sometimes, you will face multiple disconnects/service error message. So you can check if the workflow stuck in between.
If this happens, just click again the Run button and the execution will be started from where you left.
 |
| ComfyUI cloud (Original Platform) |
There are multiple fake websites falsely claiming to be the official ComfyUI platform and offering "cloud services." Recently, several unauthorized or even malicious sites have appeared, and some are ranking high on popular search engines. One of our team members was even targeted by one of these scams. So please, do not share your login credentials or join any suspicious sites. Always use the official ComfyUI platform only.
Comfy Cloud dashboard
You can upload you own workflows or third party workflows. To do this, just go to top left corner then click drop down menu then File > Open option
The real-time workflow running status can be tracked on top (tab) of the dashboard. This is in percentage with individual node name.
For Example- Here, we are testing Wan 2.2 Animate workflow. On above image you are seeing [5%] [90%] on top left corner of the tab. This means 5% has been completed as overall execution and 90% has been done for specific Sam2Segmentation node.
Once you are in, you can manage your profile settings, view API pricing, check your credit balance, and log out easily from the top-right corner of the interface.
After completion of your beta trial subscription you will be prompted with the message to charge of $20 usd/month.
As usual, the logs can be tracked from the left down corner. but, right now its in beta. Many people are facing to download the workflows output (image/videos). You can just open it in new tab by right clicking on it and then save locally.
You can learn ComfyUI by following our detailed ComfyUI guide from basic to advance where we breakdown each basic nodes that are repeatedly used in multiple workflows.
Testing Multiple workflows
We will be posting the running workflow details with input and output and execution timings.
Qwen Image 2509 Testing
Single image input
Time took- 3 seconds
Settings- Basic
Multi image input edit
Time took- 3-4 seconds
Settings- Basic
We inputted Two images (a girl with white top wearing red skirt, A torn jeans). And the workflow performed very well at the first attempt. You can see how perfectly the clothing has been replaced.
){kind=link}