Deploy your workflow as an API
Take your workflow JSON and deploy it as an API. Custom nodes, models, and Python dependencies are automatically configured for you by Builder.
ComfyUI workflow to production API in minutes.
Package all your custom nodes, LoRAs, models, and Python dependencies into an autoscaling endpoint.
How it works
Builder packages your ComfyUI workflow and environment into a reproducible build. Deploy that build to Comfy API and scale it on demand.
Share it with your team
Easily share the workflow with colleagues.
Use it anywhere
Use the API in your internal tool, application, website, workflow, and more.
Ship in minutes
Using our CLI and Skills, have your coding agent deploy everything in minutes.
Or skip the setup.
Paste this into your coding agent.
Built for teams
Easily share workflows with your team. APIs automatically scale up and down based on your team's needs.
Autoscaling
APIs scale up and down with your team or project’s needs to keep you moving fast. No engineering required.
Billing & audit logs
View every workflow run, and how much your team is spending. Debug using built-in logs from the console.
Run any workflow & model
Supports the thousands of open source models, custom nodes, and partner models available in the ComfyUI ecosystem.
CREATIVE APPS
Build the app, not just the workflow.
Deploy your ComfyUI workflows as APIs behind customer-facing products, internal tools, and automated creative pipelines. Build the experience your users need without exposing the graph.
Browse appsAt Silverside, Comfy workflows are where our creative expertise becomes infrastructure. We build production pipelines then serve them through Comfy API so our clients can run the same proven workflows at scale. It lets a small team deliver enterprise-grade creative output.Allie Wrubel, Head of Technology at Silverside AI
Comfy API pricing
Comfy API is a separate product from your Comfy Cloud plan, for teams building on Comfy Cloud programmatically. GPU time and storage are billed by the second; release, deployment, and concurrency limits apply to each plan below.
GPU
RTX PRO 6000
96 GB VRAM
$4.54/hr
957.94 credits/hr
H100 SXM
80 GB VRAM
$6.23/hr
1314.53 credits/hr
H200 SXM
141 GB VRAM
$7.71/hr
1626.81 credits/hr
B200
180 GB VRAM
$11.23/hr
2369.53 credits/hr
Billed by the GPU second
Storage
Network Storage
$0.20/GB/mo
42.20 credits
Models live on persistent network storage shared across a deployment's workers. For example: 20 GB of models = $4.00/mo + GPU time.
Total releases limit
5
Total deployments limit
2
Max worker concurrency (per deployment)
2
Total releases limit
5
Total deployments limit
2
Max worker concurrency (per deployment)
2
Total releases limit
10
Total deployments limit
5
Max worker concurrency (per deployment)
10
Total releases limit
40
Total deployments limit
20
Max worker concurrency (per deployment)
20
Enterprise limits are determined by your contract. Contact Us
00:00
DEVELOPER PLATFORMDeploy your ComfyUI workflow as a production API. Thousands of models on one platform.
Products
The AI creation engine for complete control
Over every model, every node, every step, every output.
Comfy Desktop
Run ComfyUI on your own hardware.
SEE DESKTOP FEATURESComfy Cloud
The full power of ComfyUI from anywhere.
SEE CLOUD FEATURESDeveloper Platform
Deploy and scale ComfyUI workflows.
SEE PLATFORM FEATURESComfy Enterprise
Enterprise-grade infrastructure for the creative engine inside your organization.
SEE ENTERPRISE FEATURES