# MiniMax Music 3

> Audio AI Model by MiniMax — available on Layer

High-performance MiniMax music model for complete songs up to five minutes with structure tags and seed control.

MiniMax Music 3 from MiniMax generates complete songs up to five minutes from a music description and lyrics. Control style, mood, vocals, and arrangement via structured captions (genre, BPM, key), and shape song form with structure tags such as [verse], [chorus], [bridge], and [outro]. Supports seed, guidance scale, and inference steps for reproducible, high-quality 44.1 kHz stereo WAV output.

## Specifications

| Property | Value |
|----------|-------|
| Provider | MiniMax |
| Category | Audio |
| Price | 0.04 / second per second |

---

[Try MiniMax Music 3](https://next.app.layer.ai/me/session/new?baseModelId=MINIMAX_MUSIC_3) | [All Audio Models](https://layer.ai/models/audio) | [All Models](https://layer.ai/models)