Skip to content
moonez.aiSign in
Catalog update Gemini 4 Argon & more

AI infrastructure for your next product

From API call
to possibility.

Build with Gemini, Nano Banana, Veo, and Lyria. Image, text, video, and music generation — one platform, one balance, your workflow.

Start building Read the docs

Prepaid billing. Model-level permissions. Built for server-side integration.

YOUR PRODUCTMOONEZ.AI / API
example.sh
POST /api/v1/generations
curl https://api.moonez.ai/api/v1/generations \
  -H "Authorization: Bearer $MOONEZ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3.1-flash-lite-image",
    "prompt": "A ceramic cup, studio lighting",
    "aspect_ratio": "1:1",
    "resolution": "1K"
  }'
202 AcceptedPoll the job. Receive the result.

01 / Send a requestNative API · cURL
Image, text & video Gemini-compatible API Async jobs & webhooks Usage-based billing

The model catalog

One platform.
More ways to build.

25 models for images, text, video, music, embeddings, and multimodal workflows. Exact API IDs, organized by what you want to create.

Embeddings

Map text, images, PDFs, audio and video into vectors with Gemini Embedding 2.

Per input token, by modality; no output-vector charge

Image generation

Create from a prompt or edit an existing image with the Nano Banana family.

Per image, by model and resolution

Text generation

Use Gemini request formats for text responses, streaming, and token counting.

Per input and output token

Video generation

Generate short videos with Veo through asynchronous jobs or the Interactions API.

Per second, by model and resolution

Multimodal

Access Gemini Omni through the Gemini-compatible generateContent API.

Per input and output token

Music generation

Create tracks with Lyria or steer a live audio stream with Lyria RealTime.

Per completed track; RealTime per second of delivered audio

Availability depends on your API key permissions and the model’s provider. See access requirements

Designed for developers

Your workflow.
Not another rewrite.

Choose the integration that fits your product. Both APIs use the same key and prepaid balance.

01 / Native API

Submit. Track. Deliver.

A job-based interface for applications that need asynchronous generation and delivery callbacks.

POST /api/v1/generations
  • Job status and result URLs
  • Polling or webhook delivery
  • Request metadata and idempotency keys
Explore the native API
02 / Gemini-compatible API

Keep your client. Change the URL.

Use Gemini request and response formats with a compatible client or the official Python SDK.

/api/gemini/v1beta
  • JSON responses and SSE streaming
  • generateContent and countTokens
  • Key-scoped model discovery
Open the SDK guide
  1. 01

    Create your API key

    Sign in, add credit, enable the models you need, and allow your server’s IP in the dashboard.

  2. 02

    Send your first request

    Use the native jobs API or a Gemini-compatible client with your Moonez.ai key.

  3. 03

    Bring results into your product

    Poll native jobs, receive JSON or SSE from supported models, or stream live music over WebSocket.

A balance, not a subscription

Know what
you’re building on.

Fund your account and use it across models. See the rates assigned to your account before you integrate.

View account pricing Sign-in required
Images
Per image, based on the model and output resolution.
Text & Omni
Input and output tokens are priced separately.
Veo video
Per second, based on the model and video resolution.
Balance & history
Track your available balance and transactions in the dashboard.

Before you build

A few useful
answers.

Visit the documentation
Which models can I use?

The catalog above includes every active platform model, including Gemini 4 Argon, Gemini 3.8 Flash, and Nano Banana 2 Lite. Your API key must have the corresponding permissions. Provider availability can vary.

Explore the model guide
Can I use my existing Gemini client?

Yes. Point a compatible client at the Moonez.ai Gemini API and use your Moonez.ai key. The documentation includes the base URL and configuration for the official Python SDK.

Configure your client
How do I receive generation results?

The native API returns a job ID. Gemini-compatible generateContent returns JSON; supported streamGenerateContent calls return SSE. Lyria RealTime uses a bidirectional WebSocket, not SSE. Veo uses asynchronous video jobs.

Understand jobs and results
How does billing work?

Moonez.ai uses a prepaid balance. Images are billed by model and resolution, text and Omni by tokens, and Veo by video duration. Lyria is billed per completed track; RealTime per second of delivered audio. Your account shows your rates. Music calls require a configured price.

Read about billing
Where should I store my API key?

Keep it on your server, in an environment variable or a secrets manager. Never expose it in browser code or a public repository. Use separate keys and only enable the permissions each application needs.

Read the authentication guide

Less setup. More building.

Your next idea starts
with a request.

Create your account Explore the quickstart
moonez.ai