API overview
OpenAI-compatible clients use this base URL:
https://omrouter.com/v1Anthropic SDKs and some direct HTTP requests use the service address:
https://omrouter.comText endpoints
Section titled “Text endpoints”| Method | Path | Purpose |
|---|---|---|
GET |
/v1/models |
List models |
POST |
/v1/chat/completions |
OpenAI Chat Completions |
POST |
/v1/responses |
OpenAI Responses |
POST |
/v1/responses/compact |
Compact Responses |
POST |
/v1/messages |
Anthropic Messages |
POST |
/v1beta/models/{model}:generateContent |
Gemini generateContent |
Image, audio, and other endpoints
Section titled “Image, audio, and other endpoints”| Method | Path | Purpose |
|---|---|---|
POST |
/v1/images/generations |
Generate images |
POST |
/v1/images/edits |
Edit images |
POST |
/v1/audio/transcriptions |
Transcribe audio |
POST |
/v1/audio/translations |
Translate audio |
POST |
/v1/audio/speech |
Text to speech |
POST |
/v1/embeddings |
Generate embeddings |
POST |
/v1/rerank |
Rerank documents |
POST |
/v1/video/generations |
Create a video task |
GET |
/v1/video/generations/{task_id} |
Check a video task |
Authentication
Section titled “Authentication”OpenAI-compatible, image, audio, and other endpoints use:
Authorization: Bearer OMROUTER_API_KEYAnthropic uses x-api-key, and Gemini uses x-goog-api-key. See API keys and authentication for complete examples.
Request content type
Section titled “Request content type”JSON requests require:
Content-Type: application/jsonAudio and image uploads use multipart/form-data. SDKs and cURL’s --form option set the correct header automatically.