IleAI Docs

Quickstart

Account, API key, first call — in under 5 minutes.

1. Create an account

From the dashboard, or directly via the API:
bash
curl -X POST https://api.ileai.io/v1/account \
  -H "Content-Type: application/json" \
  -d '{
    "email": "[email protected]",
    "password": "un-mot-de-passe-solide",
    "name": "Votre nom"
  }'

An organization, a project ("Default Project") and an environment ("development") are created automatically — no need to configure them before your first call. A free $5 credit is also granted to the organization.

2. Generate an API key

From the dashboard (API Keys), or via the API using the environment id returned in step 1:

bash
curl -X POST https://api.ileai.io/v1/environments/{environment_id}/api-keys \
  -H "Authorization: Bearer <jeton de session>" \
  -H "Content-Type: application/json" \
  -d '{"name": "Cle de test"}'
The raw key (emk_live_... or emk_test_...) is only returned once, at creation. Save it immediately — it can never be retrieved afterwards (only its hash is stored server-side).

3. First call

Every authenticated request uses the standard header:

http
Authorization: Bearer emk_live_...
curlbash
curl https://api.ileai.io/v1/chat/completions \
  -H "Authorization: Bearer emk_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [{"role": "user", "content": "Bonjour !"}]
  }'
JavaScript / TypeScripttypescript
import { EminiAI } from "@emini-ai/sdk";

const client = new EminiAI({ apiKey: process.env.EMINI_API_KEY! });

const result = await client.chat.completions.create({
  messages: [{ role: "user", content: "Bonjour !" }],
});
console.log(result.content);
Pythonpython
from emini_ai import EminiAI

client = EminiAI(api_key="emk_live_...")

result = client.chat.completions.create(
    messages=[{"role": "user", "content": "Bonjour !"}],
)
print(result.content)

You didn't specify a model: the router automatically picks one (auto strategy, quality/cost balance). See Routing & fallback to pick a specific model or another strategy.

4. Next