Minimax Music · V1.5

Audio·minimax-music·by Minimax

MiniMax Music v1.5 turns simple text prompts into high-quality, expressive music. It offers a wide range of styles and moods, creating melodies and rhythms that feel natural and engaging.

Try it now →
API reference
Runtime (p50)
40s
Estimated price
$0.03
Call the API
prediction.sh
sh
curl -X POST \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "minimax-music-v1-5",
    "version": "0.0.1",
    "input": {
        "prompt": "[verse]  \nIn the heart of creation, where ideas ignite,  \nEachlabs is shining, taking brands to new heights.  \nFrom images to videos, the magic we design,  \nTech with a rhythm, innovation on the line.  \n\n[chorus]  \nEachlabs, the future, the flow, the vibe,  \nTurning every vision into something alive.  \nFast, fearless, building dreams that last,  \nWith rhythm and soul, we’re moving so fast.",
        "lyrics_prompt": "R&B, energetic"
    },
    "webhook_url": ""
}' \
  https://api.eachlabs.ai/v1/prediction/
Documentation8 sections
  • Overview

    minimax-music-v1.5 — Text-to-Audio AI Model

    Developed by Minimax as part of the minimax-music family, minimax-music-v1.5 transforms simple text prompts into complete, professional-grade songs up to 4 minutes long, solving the challenge of rapid music creation for creators without instruments or studios. This text-to-voice AI model excels in generating natural vocals, rich instrumentation, and structured tracks with lyrics, making it ideal for Minimax text-to-voice applications in content production. Users input a music prompt describing style, mood, and scenario—plus optional structured lyrics—and receive high-fidelity audio in formats like MP3 or WAV within 1-2 minutes.

  • Capabilities

    Complete Song Generation: Creates full musical compositions with lyrics, melody, harmony, and arrangement

    Multi-Genre Mastery: Handles diverse musical styles from traditional to contemporary genres

    Bilingual Processing: Supports both English and Chinese lyrics with appropriate musical styling

    Structural Intelligence: Automatically arranges songs with proper verse-chorus-bridge organization

    Vocal Synthesis: Generates realistic vocal performances that match lyrical content and musical style

    Instrumental Variety: Incorporates wide range of instruments including traditional and modern options

    Emotional Expression: Translates emotional descriptions into appropriate musical elements

    Cultural Authenticity: Accurately represents traditional music styles and cultural elements

  • Use cases

    Use Cases for minimax-music-v1.5

    Content creators producing YouTube videos or podcasts use minimax-music-v1.5 to generate custom background tracks; for example, input a prompt like "upbeat indie folk with acoustic guitar, soft vocals, [Verse] about chasing dreams, [Chorus] uplifting melody" to get a 2-minute song with natural flow in under 2 minutes.

    Songwriters and musicians prototype ideas via the minimax-music-v1.5 API, feeding structured lyrics and style descriptors like "punk rock energy with driving bass" to create full demos up to 4 minutes, streamlining the transition from concept to polished track.

    Marketers crafting brand jingles input scenario prompts such as "energetic electronic theme for tech startup, [Bridge] building tension," leveraging its vocal nuance and instrumental clarity for memorable, original audio assets tailored to campaigns.

    Developers building AI music generator apps integrate this model for user-facing tools, combining emotion control and section tags to enable real-time song creation from text, enhancing apps for educators teaching composition or gamers needing dynamic soundtracks.

  • Tips & tricks

    How to Use minimax-music-v1.5 on Eachlabs

    Access minimax-music-v1.5 seamlessly through Eachlabs' Playground for instant testing, API for production apps, or SDK for custom integrations. Provide a text prompt with style, mood, and optional structured lyrics using [Verse], [Chorus] tags, plus duration up to 4 minutes; receive high-quality MP3/WAV outputs with natural vocals and instrumentation in 1-2 minutes. Eachlabs delivers reliable, scalable access to this Minimax powerhouse.

    ---
  • Technical spec

    What Sets minimax-music-v1.5 Apart

    minimax-music-v1.5 stands out in the text-to-audio landscape with its support for detailed song structures using tags like [Verse], [Chorus], and [Bridge], enabling precise control over composition that many competitors lack. This allows users to craft full songs with logical flow, from intro to outro, maintaining musical coherence. It also delivers natural vocals and multi-instrument layers up to 4 minutes, surpassing shorter-clip models while offering rapid generation in 60-120 seconds.

    • Enhanced lyrical section control: Specify [Intro], [Verse 1], [Chorus] in prompts for structured songs, enabling songwriters to prototype demos with professional arrangement.
    • Rich style and emotion customization: Describe genres like indie pop or electronic with mood tags, producing expressive tracks with clear vocals and transients—ideal for minimax-music-v1.5 API integrations.
    • Multi-format output up to 4 minutes: Supports MP3, WAV, FLAC with customizable duration, perfect for video soundtracks or podcasts without length limitations of basic generators.
  • Things to be aware of

    Beginner Compositions

    • Simple Pop Song: "[verse] Walking down the street, feeling so free [chorus] This is my moment, just let me be"
    • Acoustic Ballad: "[verse] Quiet nights and starlit skies [chorus] In your eyes I see forever"
    • Upbeat Dance: "[verse] Move your body to the beat [chorus] Dance like nobody's watching"
    • Country Folk: "[verse] Old dirt roads and summer days [chorus] Take me home to simpler ways"

    Advanced Musical Concepts

    • Genre Fusion: Combine classical orchestra with modern electronic elements
    • Cultural Blending: Mix traditional Chinese instruments with Western pop structures
    • Complex Arrangements: Create multi-section songs with intro, multiple verses, bridge, and outro
    • Emotional Journeys: Develop songs that progress from melancholy verses to uplifting choruses

    Professional Content

    • Commercial Jingles: Short, catchy musical phrases for brand recognition
    • Film Scoring: Dramatic instrumental pieces for video content and presentations
    • Theme Development: Create recurring musical themes for series or ongoing projects
    • Mood Music: Generate ambient and atmospheric music for various professional contexts

    Experimental Projects

    • Instrumental Focus: Create purely instrumental pieces highlighting specific instruments
    • Vocal Harmony: Develop songs featuring complex vocal arrangements and harmonies
    • Tempo Variations: Experiment with songs that change tempo throughout the composition
    • Cultural Exploration: Generate authentic traditional music from various global cultures

    Educational Content

    • Music Theory Examples: Create songs demonstrating specific musical concepts and structures
    • Historical Styles: Generate music in the style of different historical periods
    • Language Practice: Produce songs in different languages for pronunciation and vocabulary practice
    • Rhythm Studies: Create compositions focusing on specific rhythmic patterns and time signatures
  • Key considerations

    Lyric Length: Stay within 10-600 character limit for lyrics to ensure quality output

    Style Clarity: Vague style descriptions may result in generic or unpredictable musical arrangements

    Cultural Context: Traditional music styles may require specific cultural or regional terminology

    Genre Mixing: Complex multi-genre requests may produce inconsistent musical results

    Tempo Specifications: Include specific tempo or rhythm descriptions for better control

    Vocal Style: Specify desired vocal characteristics (male/female, age, energy level)

    Content Guidelines: Follow appropriate content policies for lyrical material


    Legal Information for Minimax Music V1.5

    By using this Minimax Music V1.5, you agree to:

  • Limitations

    Duration Constraint: Maximum output length limited to 240 seconds (4 minutes)

    Lyric Length: Restricted to 600 characters maximum for input lyrics

    Language Limitation: Currently supports only English and Chinese languages

    Style Mixing: Very complex multi-genre combinations may produce inconsistent results

    Real-time Generation: Not suitable for live performance or real-time music creation

    Personalization: Cannot learn individual user preferences or create custom vocal styles

    Copyright Restrictions: Cannot reproduce copyrighted melodies or lyrics accurately

    Technical Complexity: Very specific music theory requests may not be interpreted correctly


    Output Format: MP3

Related models

4 models
Google · Text to Speech AI model preview
Google · Text to SpeechGoogle
Inworld TTS 1.5 AI model preview
Inworld TTS 1.5Inworld
Kokoro 82M AI model preview
Kokoro 82MKokoro
Mureka Region Edit Song · Music Editing AI model preview
Mureka Region Edit Song · Music EditingMureka
* FAQ

About Minimax Music · V1.5

01 / 03

What is MiniMax Music v1.5?

MiniMax Music v1.5 is an AI music generation model by MiniMax that creates high-quality original music tracks from text prompts. It supports diverse genres and vocal styles, producing complete compositions for use in video content, games, podcasts, and creative media projects.