Back to Nodes
Minimax Music 2.5

Minimax Music 2.5

Official

Full-length song generation with vocals, lyrics, and rich instrumentation from text guidance.

Nodespell AI
AI / Audio / Minimax

Full-length song generation with vocals, lyrics, and rich instrumentation from text guidance.

Model Overview

Minimax Music 2.5 is a text-guided music generation model that creates complete songs with vocal and instrumental structure. It supports lyric-driven composition and style/mood prompting for production-ready draft outputs.

Best At

  • Generating complete songs from lyrics and style prompts.
  • Producing vocal-forward tracks with coherent arrangement sections.
  • Iterating quickly on genre, mood, and storytelling direction.

Limitations / Not Good At

  • Output quality is highly dependent on lyric quality and stylistic specificity.
  • Very niche production techniques may still require manual DAW post-processing.
  • It is not designed for speech synthesis, transcription, or sound-effect-only workflows.

Ideal Use Cases

  • Song ideation and rapid demo generation.
  • Draft background tracks for videos, short films, and social content.
  • Exploring multiple lyrical and style directions before full production.

Input & Output Format

  • Input: lyrics (required), optional style prompt, and output controls (audio_format, bitrate, sample_rate).
  • Output: Generated song audio file URI.

Performance Notes

  • Best results come from structured lyrics and explicit style guidance.
  • Audio format and quality parameters can be tuned for downstream editing or delivery.
Inputs (2)

Lyrics

String

Lyrics for the song. Use \n to separate lines. Supports section tags like [Intro], [Verse], [Pre Chorus], [Chorus], [Bridge], and [Outro]. 1-3500 characters.

RequiredMulti InputMin: 0Max: 100

Prompt

String

Describe style, mood, genre, and arrangement direction. Example: 'melancholic cinematic pop with warm female vocal'.

Multi InputMin: 0Max: 100
Parameters (5)

Prompt

String

Describe style, mood, genre, and arrangement direction. Example: 'melancholic cinematic pop with warm female vocal'.

Default:

Lyrics

String

Lyrics for the song. Use \n to separate lines. Supports section tags like [Intro], [Verse], [Pre Chorus], [Chorus], [Bridge], and [Outro]. 1-3500 characters.

Required
Default:

Sample Rate

Number

Sample rate for the generated music.

Default: 44100

Bitrate

Number

Bitrate for the generated music.

Default: 256000

Audio Format

String

Output audio format.

Default: mp3
Outputs (1)

Output

Inferred

Output

Nodespell Team

Type

Node

Status

Official

Package

Nodespell AI

Category

AI / Audio / Minimax

Input

Text

Output

Audio

Keywords

Music GenerationStyle ControlLength ControlPrompt Conditioning
Use in Workflow