Skip to main content
HiLinkup exposes OpenAI-compatible endpoints and selected provider-style compatible endpoints. Most OpenAI-compatible clients only need a new API key, base URL, and model ID. Claude-style and Gemini-style clients must use their own endpoint paths and auth headers.

​
Base URLs and full endpoints

When an SDK asks for a base URL, use:
When a client asks for a full endpoint, use the protocol-specific path:
Do not interchange a base URL and a full endpoint. SDKs normally append the resource path, so using a full endpoint as a base URL can create duplicated paths.

​
Model IDs and providers

GET https://api.hilinkup.com/v1/models identifies the provider in owned_by and returns id values in provider/model format. Sending aliyun/kimi-k3 explicitly uses the aliyun provider. Sending only kimi-k3 uses the platform’s current default provider setting.
For predictable provider routing, copy the complete data[].id returned by the model list. Available providers and the platform default can change.

​
Authentication

OpenAI-compatible endpoints:
Claude Messages-compatible endpoints:
Gemini-compatible endpoints:

Responses

Chat Completions

Claude Messages

Gemini generateContent

Models

Embeddings

Image generation

Image editing

Last modified on August 12, 2026