AI Music Generator

Back to pipelines

Create your first track

Describe the sound you imagine, or try an example.

Complete to generate

Mood / Genre

Describe genre, mood, instruments, tempo, key, structure, and the track's intended use.

Runs on

4 models

Google
- Lyria 3 Pro
- Lyria 3.5
ElevenLabs
- Eleven Music v1
- Eleven Music v2.5

Examples

Audio

Eleven Music v2.5

Pyra's Violet Wings (Eleven Music v2.5)

Eleven Music v2.5. The same prompt and length as Pyra's original theme, so you can hear what the newer model changes: exactly 40 seconds of 48 kHz stereo.

Reproduce

How to use Music Generator

A compact decision guide for placing this pipeline in a workflow.

Best for

Background scores for documentaries, podcasts, and videos
Mood-setting ambient music, cinematic underscore, lo-fi, orchestral
Genre-specific tracks anchored to a reference style (EDM, trailer, corporate uplifting)

Tips

Be specific about mood: 'dark cinematic orchestral, slow tension build, suspenseful atmosphere' not 'dramatic'
Request duration ~10% longer than video to ensure full coverage
For background scores: request instrumental only, no vocals

Music Generator API

Call this pipeline from your own code. One request dispatches a run; the model is an input field, not a separate endpoint.

Get an API token

inputs: 4
required: 2
Pricing: 2.5 to 22.6 credits

Language

TypeScript Python Go CLI

bun add @pipe2-ai/sdk

Run a pipeline

import { createClient } from "@pipe2-ai/sdk";

const client = createClient(process.env.PIPE2_TOKEN!);

const { run_pipeline } = await client.RunPipeline({
  pipeline_slug: "music-generator",
  input: {
    prompt: "Warm lo-fi beat with a soft Rhodes piano, relaxed and unhurried",
    duration_sec: 60,
    model: "eleven-music-v2-5",
  },
});

pip install pipe2

Run a pipeline

import asyncio
import os

from pipe2 import Pipe2Client

client = Pipe2Client(token=os.environ["PIPE2_TOKEN"])

run = asyncio.run(client.run_pipeline(
    pipeline_slug="music-generator",
    input={
        "prompt": "Warm lo-fi beat with a soft Rhodes piano, relaxed and unhurried",
        "duration_sec": 60,
        "model": "eleven-music-v2-5",
    },
))

go get github.com/pipe2-ai/sdk-go

Run a pipeline

client := pipe2.NewClient(os.Getenv("PIPE2_TOKEN"))

input := json.RawMessage(`{
  "prompt": "Warm lo-fi beat with a soft Rhodes piano, relaxed and unhurried",
  "duration_sec": 60,
  "model": "eleven-music-v2-5"
}`)

run, err := pipe2.RunPipeline(context.Background(), client, "music-generator", input)
if err != nil {
    log.Fatal(err)
}

brew install pipe2-ai/tap/pipe2

Run a pipeline

pipe2 pipelines run \
  --pipeline music-generator \
  --input-json '{"prompt": "Warm lo-fi beat with a soft Rhodes piano, relaxed and unhurried", "duration_sec": 60, "model": "eleven-music-v2-5"}' \
  --wait

Per page

Input	Type	default	Description
`duration_sec`required	integer	default60	Requested track length in seconds. Model output can vary slightly.
`prompt`required	string	default—	Describe genre, mood, instruments, tempo, key, structure, and the track's intended use.
`model`	enum	defaultauto	Music engine to use. Leave unset to let Pipe2 choose. auto · eleven-music-v2-5 · lyria-3-5 +2 `autoeleven-music-v2-5lyria-3-5eleven-music-v1lyria-3-pro`
`vocals`	boolean	defaulttrue	Turn off to require or request an instrumental track. When enabled, the selected model may include vocals based on the prompt.

Point any MCP client at pipe2 and your agent gets these tools. It finds this pipeline, reads the same schema above, then runs it.

list_pipelines
get_pipeline_schema
run_pipeline
get_pipeline_run_status
request_upload

https://mcp.pipe2.ai/mcp

{
  "mcpServers": {
    "pipe2ai": {
      "url": "https://mcp.pipe2.ai/mcp",
      "headers": { "Authorization": "Bearer YOUR_TOKEN" }
    }
  }
}

MCP setup

Full API reference CLI reference

Recipes using this pipeline

Show all recipes

One prompt → a vertical AI dance reel

Generate a labeled dance-move grid, use it as the choreography reference for a music-synced dance, then combine both into a vertical reel.

6 pipelines one CLI command

Open recipe
1. 1 Image Generator
2. 2 Music Generator
3. 3 Image Motion
4. 4 Video Generator
5. 5 Video Reel
6. 6 Watermark
Open recipe

Articles about this pipeline

Show all articles

Frequently Asked Questions

What audio format is the output?

High-quality MP3, ready to drop into any video editor or browser. Files land in your asset library for immediate use in the Video Reel pipeline.

How long can the track be?

From 10 seconds up to 5 minutes. Eleven Music covers the full range, while Lyria caps at 3 minutes. Pick the requested duration on the slider; generated length can vary slightly.

Are tracks instrumental or do they have vocals?

Vocals are allowed by default. Turn them off to require or request an instrumental bed under narration, where singing would compete with the speaking voice. When vocals are enabled, the model still follows the prompt and may choose an instrumental arrangement.

Can I use generated music commercially?

Yes. All generated music is cleared for commercial use. Check the usage policies for platform-specific restrictions.

How specific should my prompt be?

Very. Instead of 'happy music', try 'upbeat acoustic guitar with light percussion, 120 BPM, summer road trip feeling'. Naming instruments, tempo, and emotional arc produces much better results.