Skip to content

Developer Guide Overview

Applies to: Developers Updated: 2026-07-02

CloudRouter offers an API that is fully compatible with the OpenAI / Anthropic / Gemini ecosystems. If your code already uses the OpenAI or Claude SDK, you can usually switch to CloudRouter by changing just two lines of config (Base URL + API Key). One integration lets you call all the world's mainstream models.


1. Integration takes just three steps

① Get an API Key      → Created by an Organization or Subuser on the console "API Keys" page
② Set the Base URL    → Point requests at CloudRouter
③ Send a request      → Call using the familiar OpenAI/Anthropic format

For detailed steps, see Quickstart: Your First Call in 5 Minutes.


2. Core concepts

Understand the few concepts below and you understand the entire call chain:

ConceptWhat it isAnalogy
API KeyYour identity credential, starts with sk-Access card
Base URLCloudRouter's API addressServer street address
Model call nameThe value you put in the request's model fieldWhich dish you order
GroupA labeling dimension for same-named models, used for disambiguation and routingYour menu section

For the model call name, v2 offers two forms:

  • Bare model name, such as claude-sonnet-4-6 or gpt-4o: routes to the routing group bound to this Key (same behavior as the legacy console).
  • With fingerprint, such as 5MHXZWKA/gpt-4o: routes to a specific group by "calling it out" via the fingerprint.

For both forms, copying from the console's Call Guide page is the safest approach. For the detailed mechanics, scenario selection, and gateway routing rules, see Call Guide & Dual Routing.

Which groups / models you can call depends on the range of models available to your account (configured per organization by the platform). If you can't reach a certain model, first confirm your available range — see Admin Manual · Regions & Model Scope.


3. Protocol formats

CloudRouter is compatible with several mainstream protocols; pick whichever you're used to:

ProtocolEndpointBest forExample docs
OpenAI format/v1/chat/completionsUsing the OpenAI SDK, most frameworksAPI Basics
Anthropic format/v1/messagesUsing the Anthropic SDK, calling ClaudeAPI Basics
Gemini format/v1beta/models/*Using the Google Gemini SDKAPI Basics

All formats call the same set of models; they differ only in the field structure of the request/response. The same model can be called with either the OpenAI format or the Anthropic format.


4. Pick a doc by your scenario

Your situationRecommended path
First time using it, want to get something runningQuickstart
Want to figure out exactly what to put in the model fieldCall Guide & Dual Routing
Have existing OpenAI/Claude code, want to migrateSDK Integration
Using tools like Claude Code / CursorThird-party Tool Integration
Want to learn about request parameters and streamingAPI Basics
Hit an errorErrors & Troubleshooting
Model occasionally outputs Korean/JapaneseClaude Language Drift Note
Want to quickly look up the answer to a questionDeveloper FAQ

5. Advanced usage

The following capabilities are already supported; dedicated docs are coming soon:

  • Prompt Caching (supported, docs in progress)
  • Tool Calls / Function Calling (supported, docs in progress)
  • Multimodal (image input) (supported, docs in progress)
  • Structured Output (supported on some models) (supported, docs in progress)

Contact us