Create a key
Sign in and create a workspace API key. Save it securely.
Developer quickstart · 5 minutes
Make your first request with the tools you already use.
Create a workspace key, read the enabled model catalog, then send a standard chat-completions request from your server. The API uses a familiar OpenAI-compatible format.
- Keep the ngm_ key in a server-side secret
- Choose a model ID returned by GET /v1/models
- Inspect usage and credit balance in your workspace
first-request.shcURL
curl https://api.ngamia.cc/v1/chat/completions \
-H "Authorization: Bearer $NGAMIA_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"MODEL_FROM_CATALOG","messages":[{"role":"user","content":"Habari"}]}'Get the model identifier from GET /v1/models.
How it comes together
Your first request, without a framework detour.
The same endpoint works with curl or an OpenAI-compatible SDK. Start with a model ID from the live catalog.
Choose a model
Fetch GET /v1/models; do not guess or hard-code an unavailable ID.
Send a prompt
POST to /v1/chat/completions with your model and messages.
The example on this page is illustrative. Replace MODEL_FROM_CATALOG with an ID returned for your key.