Build with AI, without the complexity.
One API for the models you need.
QUICKSTARTFull quickstart
- 01Create an accountEmail address, no sales call.
- 02Add creditsSpent per request. No monthly minimum.
- 03Create an API keyOne per project. Shown once, so copy it.
- 04Make your first requestOne address, one key.
Let an agent wire it up
Working in Claude Code, Codex or Gemini CLI? Paste this. It points the agent at /llms.txt . The whole integration written for a machine: base URL, auth header, where model slugs come from, every error code and which of them are worth retrying.
Read https://routehook.ai/llms.txt and wire this project up to
Routehook exactly as it specifies. Use my ROUTEHOOK_API_KEY from the
environment, pick a model slug from GET /v1/models rather than guessing one,
and only add retry logic for 429 and 503.
Your first request
Every endpoint takes the same bearer token and returns JSON. Swapping models is a change to one string.
curl https://api.routehook.ai/v1/chat/completions \
-H "Authorization: Bearer $ROUTEHOOK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "your-model",
"messages": [
{ "role": "user", "content": "Hello" }
]
}'
The response
{
"id": "chat_5f81c0",
"object": "chat.completion",
"model": "your-model",
"choices": [
{
"index": 0,
"message": { "role": "assistant", "content": "Hello. How can I help?" },
"finish_reason": "stop"
}
],
"usage": { "input_tokens": 9, "output_tokens": 8, "total_tokens": 17 },
"cost": 0.000103,
"reference_cost": 0.000245
}
The last two fields are on every response: what the call cost you, and what the same call costs at the provider’s published price.
Explore
QuickstartAccount to first response, in order.AuthenticationBearer tokens, x-api-key, and where a key belongs.API ReferenceEvery route, its method and whether it is callable.Model OverviewModel ids, modalities and live rates.Error HandlingStatus codes, the error envelope and retries.SDKsThe clients that already speak these shapes, and the base URL each wants.