Client Compatibility
Starrise-api is fully compatible with the OpenAI API format. Any client, SDK, or tool that supports OpenAI can connect b
Configuration
| Setting | OpenAI Default | Starrise-api |
|---|---|---|
| Base URL | https://api.openai.com/v1 | https://api.starrise.ai/v1 |
| API Key | OpenAI API key | Starrise-api token (sk-...) |
Sign in to the console and copy your API address (Base URL) from the home page. Paste it into your SDK or third-party tool as the Base URL / API Host. Example:
https://api.starrise.ai/v1When using the official OpenAI SDK, make sure the Base URL ends with /v1.
Official SDKs
Python
pip install openaifrom openai import OpenAI
client = OpenAI(
base_url="https://api.starrise.ai/v1",
api_key="sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
)Node.js
npm install openaiimport OpenAI from 'openai'
const client = new OpenAI({
baseURL: 'https://api.starrise.ai/v1',
apiKey: 'sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'
})OpenAI-Compatible Endpoints
These routes use Authorization: Bearer sk-... (or sk- prefix without Bearer in some tools).
| Endpoint | Method | Description |
|---|---|---|
/v1/chat/completions | POST | Chat completions (streaming supported) |
/v1/completions | POST | Legacy text completions |
/v1/responses | POST | OpenAI Responses API |
/v1/responses/compact | POST | Responses compaction |
/v1/embeddings | POST | Text embeddings |
/v1/rerank | POST | Document reranking (Jina-style) |
/v1/images/generations | POST | Image generation |
/v1/images/edits | POST | Image editing |
/v1/edits | POST | Legacy image edit |
/v1/audio/transcriptions | POST | Speech-to-text |
/v1/audio/translations | POST | Audio translation |
/v1/audio/speech | POST | Text-to-speech |
/v1/moderations | POST | Content moderation |
/v1/models | GET | List models (OpenAI format) |
/v1/models/:model | GET | Retrieve model |
/v1/realtime | GET (WebSocket) | OpenAI Realtime API |
Not implemented
/v1/images/variations, /v1/files/*, and /v1/fine-tunes/* return not-implemented on this gateway.
Claude (Anthropic) Endpoints
| Endpoint | Method | Auth |
|---|---|---|
/v1/messages | POST | Authorization: Bearer sk-... or x-api-key + anthropic-version |
Model list with Anthropic headers: GET /v1/models with x-api-key and anthropic-version.
Gemini Endpoints
| Endpoint | Method | Auth |
|---|---|---|
/v1beta/models | GET | Authorization, x-goog-api-key, or ?key= |
/v1beta/models/*path | POST | Same as above |
/v1beta/openai/models | GET | OpenAI-style model list |
/v1/models/*path | POST | Gemini relay under OpenAI path prefix |
Midjourney (/mj)
Requires token via Authorization or mj-api-secret header.
| Endpoint | Method | Description |
|---|---|---|
/mj/submit/imagine | POST | Create image job |
/mj/submit/change | POST | Upscale / variation |
/mj/submit/describe | POST | Describe image |
/mj/submit/blend | POST | Blend images |
/mj/task/:id/fetch | GET | Fetch task status |
/mj/image/:id | GET | Fetch result image |
Suno (/suno)
| Endpoint | Method | Description |
|---|---|---|
/suno/submit/:action | POST | Submit music generation |
/suno/fetch | POST | Batch fetch tasks |
/suno/fetch/:id | GET | Fetch single task |
Third-Party Tools
| Tool | Configuration |
|---|---|
| ChatGPT Next Web | Settings → API Host → https://api.starrise.ai |
| LobeChat | API proxy URL + token |
| Cursor / Cline | Override OpenAI Base URL |
| LangChain | OpenAI(base_url=..., api_key=...) |
| Dify | OpenAI-compatible provider |
Environment Variables
OPENAI_API_BASE=https://api.starrise.ai/v1
OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxTroubleshooting
| Issue | Solution |
|---|---|
401 Unauthorized | Check token; use Bearer sk-... |
403 Forbidden | Model/IP restriction or banned user |
404 Model not found | Call GET /v1/models or check console pricing |
| Connection timeout | Ensure base URL ends with /v1 for OpenAI SDK |
| SDK still hits OpenAI | Set base_url / baseURL explicitly |