Developer API

AI Music Generator API for Developers

Generate original instrumental tracks or vocal songs from a style description and your lyrics. Start with cURL or TypeScript, or connect Cursor, Claude Code, and Codex through MCP.

Available on Pro and Ultra plans with credit-based usage.

cURL quickstart

Terminal

curl https://www.lacuna.fm/api/v1/music/generations \
  -H "Authorization: Bearer $LACUNA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "style": "lofi hip hop, mellow piano, 70 bpm",
    "title": "Late Night Study",
    "instrumental": true,
    "model": "aether"
  }'
POST /api/v1/music/generationsCreate API key
Built for developers and agents

An AI music generation API that fits your workflow

Move from one request to a hosted track through direct code, an SDK, or the coding agent you already use.
Lacuna API request producing a complete AI-generated music track with hosted audio

Generate complete tracks from one request

  • Describe the style and title you need
  • Choose instrumental mode or provide your own lyrics
  • Receive complete tracks with hosted audio and metadata
Cursor, Claude Code, and Codex using Lacuna documentation to add AI music generation

Connect Cursor, Claude Code, and Codex

  • Run the official lacuna-mcp server directly from npm
  • Add one API key to your coding agent's MCP configuration
  • Ask for music in plain language and receive a hosted track
Set Up Your Coding Agent
TypeScript SDK and REST API code with typed music generation fields and errors

Use REST, TypeScript, CLI, or MCP

  • Call the REST API from any backend language
  • Use typed create, retrieve, and waitFor methods on Node.js 18+
  • Verify webhook signatures with the official SDK
AI music generation task moving from queued to processing to ready with a signed webhook

Build around a clear asynchronous lifecycle

  • Create a task and poll it or call waitFor
  • Receive signed job.completed and job.failed webhooks
  • Get automatic credit refunds when generation fails
Choose your integration

Start from the interface that fits your stack

The same Lacuna Music API is available through direct HTTP calls, a typed SDK, command-line tools, and an MCP server for Cursor, Claude Code, Codex, and other coding agents.

REST API

Call the HTTP endpoints from any language and inspect every request and response in the OpenAPI reference.

View API Reference

TypeScript SDK

Create and retrieve generations with typed methods, waitFor polling, structured errors, and webhook verification.

Read the Quickstart

Command Line

Generate and monitor tracks from a terminal with the Lacuna Toolkit CLI.

View CLI Setup

MCP for AI Agents

Give compatible coding agents access to music generation through the Lacuna MCP server.

Use with AI Agents
Use cases

Two broad workflows, one generation API

Use complete generated tracks inside interactive products or automate music production behind your content and agent workflows.

Music application and game prototype using AI-generated instrumental tracks from Lacuna
01

Music apps and games

Let users generate original tracks, score prototypes, and create music for interactive scenes without leaving your product.

Automated video, podcast, and coding agent workflow receiving generated music from Lacuna
02

Content and agent workflows

Automate original music for videos, podcasts, campaign assets, playlists, and tasks initiated by your coding agent.

FAQ

AI Music API questions

Current access, generation, billing, and integration details.

The API creates complete instrumental tracks or vocal songs. Every request includes a title and style description; vocal generations also require lyrics supplied by your application.

No. For a vocal generation, your application supplies the lyrics. Use Lacuna's separate lyrics tools if you need to create lyrics before sending the music request.

No. Creating a generation returns an asynchronous task. You can poll its status, use the TypeScript SDK's waitFor helper, or subscribe to signed job.completed and job.failed webhooks.

Create an API key in your Lacuna profile, keep it in a secrets manager, and send it as a Bearer token. Keys can be rotated or revoked from the same dashboard.

Yes. Install the official lacuna-mcp server, add your Lacuna API key to the agent's MCP configuration, and ask for tracks in natural language. The same tools work in Cursor, Claude Code, Codex, and other MCP-compatible clients.

The Music API is available on Pro and Ultra plans. Usage is credit-based; current plan details are listed on the pricing page.

The creation endpoint allows 60 requests per minute per API key. Default concurrent generation limits are 10 tasks on Pro and 20 on Ultra; account-specific overrides may apply.

Credits are deducted when the task is created and refunded automatically if the generation fails. Typed errors and webhook events let your application handle the failure explicitly.

Songs generated during an active Pro or Ultra subscription are covered by Lacuna's commercial license, and those commercial rights continue after the subscription ends.

Build with Lacuna

Make your first AI music API request

Create an API key, copy a ready-to-run cURL command, or connect Cursor, Claude Code, and Codex through the Lacuna MCP server.

Get API KeyView Pricing