One REST endpoint, any AI provider. OpenRouter, Groq, Google Gemini, or any custom OpenAI-compatible server — switchable per request, pluggable in minutes.
checking API…
| Method | Path | Auth | Description |
|---|---|---|---|
| GET | /api/v1/health | public | Liveness probe |
| GET | /api/v1/providers | key* | List providers, models & configuration status |
| POST | /api/v1/chat | key* | Chat completion via any configured provider |
* key = required only when API_SECRET_KEY is set on the server (header: X-API-Key)
copy# Chat (default provider)
curl -X POST https://YOUR-DOMAIN/api/v1/chat \
-H "Content-Type: application/json" \
-d '{"message": "What is the meaning of life?"}'
copy# Pick a provider + multi-turn
curl -X POST https://YOUR-DOMAIN/api/v1/chat \
-H "Content-Type: application/json" \
-d '{
"provider": "openrouter",
"messages": [
{"role": "user", "content": "Hi!"},
{"role": "assistant", "content": "Hello!"},
{"role": "user", "content": "Tell me a joke"}
],
"temperature": 0.8
}'