RemoteGPU Docs

RemoteGPU docs are organized around three ways to run work on the platform: guided applications, direct API calls, and Kubernetes resources.

Choose a starting point

Application

Use Application when you want a guided console workflow and do not want to manage Kubernetes resources or call the API directly.

Start with Application overview, or go directly to Comfy to run a hosted workspace.

Inference API

Use Inference API when you want to generate images from your application, backend, script, or automation with HTTP requests.

Start with Inference API overview, then use Image for the first request flow.

Kubernetes

Use Kubernetes when your team wants namespace access and Kubernetes-native resources such as Deployment, Job, Service, CronJob, and Ingress.

Start with Kubernetes overview. Use API keys when you need kubectl access.

Core documentation paths

Application:

Application overview: choose hosted applications instead of API or Kubernetes workflows
Comfy: run and manage a hosted Comfy workspace

Inference API:

Inference API overview: understand direct API access and how it fits with the console
Image: send image-generation requests and track results
API keys: create keys for inference and Kubernetes exec access

Kubernetes:

Kubernetes overview: understand how console and kubectl work together on RemoteGPU
Kubernetes deployments: run workloads in a namespace
Kubernetes services: expose workloads inside the cluster
Kubernetes ingresses: publish services with supported ingress rules
Kubernetes jobs: run one-off batch work in a namespace
Kubernetes cronjobs: schedule recurring batch work in a namespace

Service endpoints

Endpoint	Purpose
console.remotegpu.ai	Customer console
`api.remotegpu.ai`	RemoteGPU API host
docs.remotegpu.ai	Customer documentation

RemoteGPU Docs ​

Choose a starting point ​

Application ​

Inference API ​

Kubernetes ​

Core documentation paths ​

Service endpoints ​

Read next ​