Provider Routing
ShareKiteRouter exposes an OpenAI-compatible API to customers. Admins can map one customer-facing model to multiple upstream provider models, ordered by priority.
How model routing works
- The client sends
model: "gpt-4o-mini"to/api/v1/chat/completions. - KiteRouter authenticates the key, checks credits, and loads active provider mappings for that model.
- KiteRouter rewrites the public model ID to the selected upstream model ID.
- The request body is forwarded to the provider's OpenAI-compatible endpoint.
- If the first provider fails, KiteRouter can try the next active mapping.
Public model vs upstream model
| Name | Example | Who sees it |
|---|---|---|
| Public model ID | gpt-4o-mini | Your customers and API users |
| Upstream model ID | openai/gpt-4o-mini or provider-specific ID | Admin/provider mapping only |
| Provider base URL | https://api.openai.com/v1 | Admin/provider configuration |
Warning: Send OpenAI-compatible request bodies. Provider-specific differences belong in admin provider configuration and upstream compatibility.