Lumen Gateway API

formerly Empire Gateway; also available as the EAS Console

Free $0 1k req/mo · Indie $19/mo · Pro $89/mo · Enterprise $449/mo

Gateway plans are separate from Lumen Ask/Agent subscriptions.

Two endpoints, two jobs. The Answer API gives you Lumen's answer engine — search, synthesis, citations and agent mode — in one call. The Router gives you an OpenAI-compatible /v1 that picks the cheapest model able to answer each request.

1 · Answer API

Authenticate with a Lumen API key as a bearer token. Signed-in browser sessions work too, so you can try a call from the console while logged in.

curl -X POST https://lumen.eliteaiempire.com/api/v1/answer \
  -H "Authorization: Bearer $LUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "q": "What changed in EU AI Act enforcement this quarter?",
    "mode": "deep",
    "model": "auto"
  }'

Request fields

Field	Type	Meaning
`q`	string, required	The question. Two characters minimum.
`mode`	`quick` · `deep` · `agent`	`quick` answers in one pass; `deep` researches more widely before answering; `agent` runs the step-by-step loop with tools. Defaults to `quick`.
`model`	string	A roster id from /api/models, or `auto` to let the cost-router choose. Defaults to `auto`.
`tier`	string	Requested capability tier. Your key's plan is the ceiling.
`messages`	array	Prior turns as `{role, content}`, OpenAI-style — the last 12 are used as conversation context.
`tid`	string	Alternative to `messages`: load an existing Lumen thread as the context.
`build_mode`	boolean	Routes to the coding models and returns code-shaped output. Prefixing `q` with `/build` does the same thing.

Keys

Create and rotate your key on the account page. Keys begin with lk_. Rotating invalidates the previous key immediately.

2 · The router (OpenAI-compatible)

If you already call an OpenAI-shaped API, you do not need the Answer API at all. Point your existing client at the gateway base URL and keep your code:

from openai import OpenAI

client = OpenAI(
    base_url="https://gateway.eliteagenticsolutions.com/v1",
    api_key="<your gateway key>",
)

client.chat.completions.create(
    model="auto",              # or any id from the roster
    messages=[{"role": "user", "content": "Summarise this contract clause."}],
)

model: "auto" hands the choice to the cost-router: it classifies how hard the request is and sends it to the cheapest model that can still answer it correctly. Ask for a specific model instead and you get exactly that model.

The full quickstart for the router lives at lumen-api.eliteaiempire.com/docs.

3 · Model roster

GET /api/models returns every model the router can reach right now, its tier, whether it needs your own key, and the list of excluded model origins. It is generated from the live provider chain, so it is never out of date.

Other endpoints

Endpoint	Does
`GET /api/models`	The live model roster.
`GET /api/stats`	Public service counters.
`POST /api/vision`	Ask a question about an uploaded image.
`POST /api/imagine`	Generate an image from a prompt.
`POST /api/voice`	Transcribe audio, then answer it.
`POST /api/upload`	Attach a document to a thread for the agent to read.
`GET /healthz`	Liveness.

Limits

Free keys are capped at 50 requests a day and paid keys at 600; see pricing for the current ladder. Exceeding the cap returns HTTP 429 with the reset time — back off and retry rather than rotating keys.

Get an API key →All Lumen products →

Gateway pricing FAQ

Free $0 1k req/mo · Indie $19/mo · Pro $89/mo · Enterprise $449/mo