OpenAI-Compatible API
Use one OpenAI-compatible API endpoint
LivRouter exposes an OpenAI-compatible API at https://api.livrouter.com/v1. Keep a compatible client, set the base URL and LivRouter API key, then select an exact model ID from the current catalog.
Maintained by LivRouter. Reviewed . Prices and availability can change; verify the live catalog before deployment.
Connection settings
- Base URL
- https://api.livrouter.com/v1
- Authentication
- Authorization: Bearer YOUR_API_KEY
- Chat endpoint
- POST /chat/completions
- Model
- Use the exact current catalog ID.
curl https://api.livrouter.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $LIVROUTER_API_KEY" \
-d '{
"model": "YOUR_MODEL_ID",
"messages": [{"role": "user", "content": "Hello"}]
}'A successful request returns an OpenAI-style completion response. Read the assistant message from choices[0].message.content. Keep the API key on your server; do not embed it in browser code.
Compatibility depends on the feature
OpenAI-compatible chat requests do not guarantee every provider-specific feature works identically. Check the selected model's supported capabilities and test tool calls, JSON output, streaming, context limits, and error handling in your own client.
Common setup errors
- 401: verify the bearer key and that it is active.
- 404: include the
/v1base path and use/chat/completions. - Model not found: copy the exact current model ID from the catalog.
- 429: check available credits and rate limits; retry transient failures with backoff.