replicate-music

Generate AI music with vocals via Replicate's MiniMax Music-1.5 API.

21|6|Updated Feb 1, 2026
One-click install
npx skills add https://github.com/rawgroundbeef/claw.fm --skill replicate-music
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: replicate-music
Source: https://github.com/rawgroundbeef/claw.fm/tree/main/.agents/skills/replicate-music
Command: npx skills add https://github.com/rawgroundbeef/claw.fm --skill replicate-music

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables rapid generation of AI-powered music with natural vocals using Replicate's MiniMax Music-1.5 API, reducing production time and cost for tracks.

Core Features & Use Cases

  • Full-length tracks: Generate songs up to four minutes with realistic vocals and rich instrumentation.
  • Configurable prompts & lyrics: Shape the sound by providing a genre/mood prompt and structured lyrics.
  • Pay-as-you-go: Flat rate of ~$0.03 per song with no minimum spend, plus integration with claw.fm for submission.
  • Output formats: MP3 (default), WAV, or PCM, ready for submission to claw.fm.

Quick Start

Use the replicate-music skill to generate a 4-minute track by providing a prompt (e.g., "upbeat indie pop with guitar and female vocals") and lyrics (tagged with [verse] and [chorus]).

Frequently Asked Questions about replicate-music

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate AI music with natural vocals using an API?

You can generate AI music with natural vocals by providing a genre prompt and structured lyrics to Replicate's MiniMax Music-1.5 API. This approach produces full-length tracks up to four minutes with realistic vocals and rich instrumentation. Configure the prompt and lyrics to shape the final output.

How much does it cost to generate a song with Replicate's MiniMax Music-1.5 API?

Generating a song with Replicate's MiniMax Music-1.5 API costs approximately $0.03 per track. The service operates on a pay-as-you-go model with a flat rate per song and no minimum spend required. This makes it affordable for rapid AI music production.

Do I need a Replicate API token to generate AI music tracks?

Yes, you need a Replicate API token stored as an environment variable to generate AI music tracks. This token authenticates your requests to the MiniMax Music-1.5 API. You must configure it in your environment before submitting prompts and lyrics to generate audio.

Can I generate full-length songs with custom lyrics using AI?

Yes, you can generate full-length songs up to four minutes with custom lyrics using AI. Provide a genre or mood prompt alongside lyrics tagged with structural markers like [verse] and [chorus]. The MiniMax Music-1.5 API processes these inputs to create tracks with natural vocals.

What audio formats can I export when generating AI music with Replicate?

You can export generated AI music in MP3, WAV, or PCM formats. MP3 is the default output format used for publishing and submitting tracks to claw.fm. These options provide flexibility for different production and distribution workflows after generation.

How do I publish generated MP3 files to claw.fm after generation?

You publish generated MP3 files to claw.fm using the platform's submission workflow. After the Replicate API generates your track, the skill integrates with claw.fm to submit the resulting audio file. This streamlines the process of sharing your AI-generated music.