Developer FAQ
Applies to: Developers Updated: 2026-07-02
This is an index of high-frequency developer questions, each pointing to detailed documentation.
1. Integration and authentication
Q: What do I put in the Base URL?https://api.cloudrouter.online/v1, see API Basics
Q: Where do I create / how do I get an API Key? Both Organizations and Subusers can create Keys on the console's "API Keys" page; when a Subuser creates a Key, the routing group is limited to the Organization's authorized scope, see Quickstart
Q: 401 error? Check the Authorization: Bearer header format, see Errors & Troubleshooting
2. Models and calling
Q: What exactly do I put in the model field? Copy the call name from the console's "Call Guide", bare name or with fingerprint both work, see Call Guide & Dual Routing
Q: What's the difference between a bare name and a fingerprint? A bare name routes to the routing group bound to this Key; a fingerprint calls out a specific group, see Call Guide & Dual Routing
Q: Why can't I reach a certain model? The call name is wrong, or the group is not within your account's available range (configured per organization by the platform); confirm your available range or copy from the Call Guide
Q: What's the difference between the OpenAI format and the Anthropic format? See section 5 of API Basics
Q: How do I switch models? Change the model field, see API Basics
3. Billing and rate limiting
Q: How is it charged? Who gets deducted? Priced by the group/channel actually hit (as shown on the "Model Marketplace" or the console's "Purchase" page); real charges always deduct from the owning Organization's balance
Q: Is there a billing difference between a fingerprint and a bare name? Independent of how the call name is written; priced by the group/channel ultimately reached, see Call Guide & Dual Routing
Q: What do I do about 429 rate limiting? Reduce concurrency, use exponential backoff; distinguish rate limits (API_KEY_RATE_*, wait for the window to reset) from quota/concurrency (the exact code is as actually returned by the API), see Errors & Troubleshooting
Q: What's the concurrency cap? How do I raise it? The Organization-level concurrency cap is configured uniformly by the platform; coordinate with your owning Organization, and if adjustment is needed, contact CloudRouter
4. Stability
Q: How do I handle 502? When the model provider returns 5xx, CloudRouter maps it uniformly to 502; retry with exponential backoff, see Errors & Troubleshooting
Q: What do I do when a fingerprint call reports "group unreachable"? The group is out of your available range or currently unavailable; the platform errors directly without degrading — switch to an available group, see Call Guide & Dual Routing
Q: How do I distinguish 500 from 502? 500 is a failure of the platform itself; 502 is a failure on the model provider side, see Errors & Troubleshooting
Q: The model occasionally outputs Korean/Japanese? Model language drift, see Claude Language Drift Note
Contact us
- Email: support_cloudrouter@clouditera.com
- QQ group: 712273971
- Platform: cloudrouter.online