OctaRouter gives you one account for every major LLM: register, create an API key, then point any AI app, agent, or backend at OctaRouter's base URL + key. Swap models by changing the model ID — no new vendor sign-ups or keys.
How OctaRouter works: requests are forwarded to upstream providers in a compatible shape. Use OctaRouter's base URL and your API key; follow the official OpenAI, Anthropic, or Google docs for all request and response parameters.
Steps
- Sign up at OctaRouter and verify your email.
- Open Dashboard → API keys and create a key. Copy the secret immediately — it is shown only once.
- Open the Models page:
- Copy the model ID from the card (use it in the
modelfield or Gemini URL path). - Click a protocol badge on the card to copy the matching base URL.
- If unsure, pick OpenAI-compatible — it works for almost every model.
- Copy the model ID from the card (use it in the
- Send a request:
curl "https://api.octarouter.com/v1/chat/completions" \
-H "Authorization: Bearer sk-or-..." \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-4o-mini",
"messages": [{"role": "user", "content": "Hello"}]
}'
Replace sk-or-... with your API key and use the model ID from the Models page.
The OpenAI-compatible route (POST /v1/chat/completions) is the default for most clients. Claude and Gemini also support native protocols — see API protocols for base URLs, examples, and links to official docs.
To switch models later, keep the same OctaRouter base URL and API key; change only the model ID.