Comfy Deploy

What it does

ComfyDeploy runs ComfyUI workflows, including custom nodes and models, on hosted cloud GPUs. You can share a workflow through a link, open a simplified generated interface with prompt fields, sliders, and image uploads, or deploy the workflow as a production API. Outputs are generated images or other workflow results, delivered through the playground or API.

Teams use it to let non-technical collaborators such as artists, designers, and product managers run complex ComfyUI workflows without working with nodes. Shared environments keep the same models, nodes, and settings across team members. Workflow version history includes one-click rollback, and generation can run in parallel across GPUs such as H100, A100, and L40S. Auto-scaling reduces GPU resources to zero when idle.

It suits creative and ML teams already building in ComfyUI who need shared access or an API without managing infrastructure. You need a ComfyDeploy account and a workflow built in ComfyUI. Using the API requires connecting your own tools to the deployed endpoint. Pricing and plan details are not listed on the homepage, so check the pricing page before committing.

What it can do

• Run ComfyUI workflows with custom nodes and models in the cloud
• Share workflows through links and simplified generated interfaces
• Deploy workflows as production APIs with one click
• Auto-scaling GPU execution, including scale to zero
• Workflow version history with one-click rollback
• Parallel generation across multiple GPU types

Works with

• Private S3 integration for uploading real images to validate workflows

Example workflows

Share a workflow with non-technical teammates

Input: a ComfyUI workflow with exposed prompt and image inputs. Steps: upload the workflow, define inputs and outputs, share the generated simplified UI link. Output: teammates adjust prompts, sliders, and uploads to generate results without opening nodes.

Deploy a workflow as a production API

Input: a ComfyUI workflow with defined input and output nodes. Steps: deploy the workflow in one click, connect the endpoint to your tools, test in the playground. Output: an API endpoint whose GPU resources scale with traffic.

Before you choose

• Pricing and plan details are not listed on the homepage; check the pricing page before committing.
• The homepage states the project is now open source; review the announcement for what this means for the hosted service.
• Validating workflows with real images depends on setting up the S3 integration.

Listed under

• Coding
• Creative
• Sandboxes and infrastructure
• API-deployment
• Comfyui
• Generative-AI
• GPU-cloud
• Open-source
• Team-collaboration

Get started

Use the Get Started link on the homepage, or book a demo.

1. Build and test your workflow in ComfyUI, including custom nodes and models.
2. Upload the full workflow to ComfyDeploy and define inputs and outputs with ComfyDeploy input and output nodes.
3. Share the generated interface with your team, or deploy the workflow as an API and test it in the playground.

Pricing plans and rates are not shown on the homepage. Check app.comfydeploy.com/pricing for current terms.

We couldn’t find published prices for Comfy Deploy. See their pricing page ↗

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• ComfyDeploy - ComfyUI for teams ↗www.comfydeploy.com

Company website

comfydeploy.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Human review

Checked by a person

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.