From your first key
to your own product.

01 / ELLAMI

Quick start

1. Sign in to the console

   Open ellami.pro/identity. Your keys, balance and usage history live in one account.

2. Choose how to work

   Install Ellami IDE to work on projects, or create an API key to connect your own application.

3. Set your first task

   Describe the outcome you want and how to verify it. The agent uses the project context and built-in skills.

02 / ELLAMI

Ellami IDE

The environment brings together chats, editor, files and agent tools. Open a project folder, sign in through the console and choose a model.

This Intel macOS build is locally signed and has not been notarized by Apple. This installer does not support other platforms.

CLI for your terminal

npm install -g https://www.ellami.pro/downloads/ellami-0.3.1.tgz
ellami login
ellami ask "Inspect the project and fix the sign-in form"

Enter your Ellami key when the CLI prompts you. Do not pass the key directly on the command line.

03 / ELLAMI

Skills are already connected

The built-in library covers product design, interfaces, development and verification. The server selects a suitable set for the task and loads it automatically. There is no need to install each skill yourself.

04 / ELLAMI

Connect your application

Create a key in the API keys section of the console. Store it on your server in the ELLAMI_KEY environment variable. A new key’s secret is displayed once.

The endpoint accepts a messages array in Chat Completions format. Use ellami_model to select a model. An arbitrary model field does not change the selection.

const response = await fetch("https://www.ellami.pro/api/ellami", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.ELLAMI_KEY}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    "ellami_model": "gpt-6-luna",
    "ellami_effort": "medium",
    "messages": [
      {
        "role": "user",
        "content": "Help me write a product description"
      }
    ]
  })
});

const data = await response.json();
if (!response.ok) throw new Error(data.error || "Request failed");
console.log(data.choices[0].message.content);

import os
import json
import urllib.request

request = urllib.request.Request(
    "https://www.ellami.pro/api/ellami",
    data=json.dumps({
        "ellami_model": "gpt-6-luna",
        "ellami_effort": "medium",
        "messages": [
            {
                "role": "user",
                "content": "Help me write a product description"
            }
        ]
    }).encode(),
    headers={
        "Authorization": "Bearer " + os.environ["ELLAMI_KEY"],
        "Content-Type": "application/json"
    },
    method="POST"
)
with urllib.request.urlopen(request) as response:
    data = json.load(response)
    print(data["choices"][0]["message"]["content"])

curl https://www.ellami.pro/api/ellami \
  -H "Authorization: Bearer $ELLAMI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"ellami_model":"gpt-6-luna","ellami_effort":"medium","messages":[{"role":"user","content":"Help me write a product description"}]}'

Parameters

Response text is in choices[0].message.content. Check the HTTP status before using the result. Keep keys out of browser JavaScript and public repositories.

05 / ELLAMI

A model for your task

The current Ellami API supports the following model selections. Availability and rates are checked against the provider catalog at request time.

GPT-6gpt-6-luna · gpt-6-sol · gpt-6-astra

Claudeclaude-sonnet-5 · claude-opus-5.5

Geminigemini-3.8-flash

DeepSeekdeepseek-v4.1-flash

Qwen and the provider catalog

The AITUNNEL catalog also includes Qwen3.8 Max, Flash and Qwen3 Coder. These IDs are not currently included in Ellami’s client selection list. On a self-hosted server, configure the standard route for Qwen using ELLAMI_STANDARD_MODEL and the corresponding rates. This is server configuration, not a client-key parameter.

Current AITUNNEL catalog ↗ · Checked September 27, 2026

Not every provider capability is exposed through Ellami. These docs cover text JSON requests; streaming, image generation and video generation are not supported by this endpoint.

06 / ELLAMI

CRM connected to your product

For an existing CRM, connect your site or app server to the integration API to send contacts, deals, subscriptions and events.

1. Save the key

   Rotate the integration key in an existing CRM’s Integration section. Keep it on your server only.

2. Send data

   Send a POST to console.ellami.pro/api/crm/ingest with the Bearer key and JSON of type contact, deal, subscription or event.

07 / ELLAMI

Usage you can understand

The core model is pay-as-you-go. The console catalog includes $3, $10 and $25 amounts. The cost of work depends on the selected model and input and output tokens.

Ellami tokens are balance units. Model tokens measure processed data. These are different: one million Ellami tokens are not one million model tokens.

The server takes the provider’s reported cost, applies a 1.5 service multiplier and converts it into balance units. The service margin is included in the charge. The console shows usage history and lets you set your own key limit.

Top-up availability is shown in the console. Test credits are labeled separately and do not charge money.

08 / ELLAMI

If a request fails