MiniMax Music

Generate full songs on MiniMax Music, with your lyrics sung to a structure you control, verse by verse, or clean instrumentals in any style. Hand invideo’s agent the idea, and it shapes the structure and the brief MiniMax composes from.

Trusted by teams at
Phantom X
Wonder Studios
Google
NVIDIA
Salesforce
Visa
IBM
monday.com
Cisco
Mastercard
Meta
LinkedIn
Netflix
Ford
Hilton
Siemens
HubSpot
Phantom X
Wonder Studios
Google
NVIDIA
Salesforce
Visa
IBM
monday.com
Cisco
Mastercard
Meta
LinkedIn
Netflix
Ford
Hilton
Siemens
HubSpot

Why serious creatives choose MiniMax Music

It composes to the structure you set

MiniMax Music takes your song’s architecture as instruction: intro, verse, pre-chorus, chorus, bridge, solo, outro, marked section by section, and composes to that shape. The chorus lands where you put it, not where the model felt like putting it.

The lyrics you write are the lyrics it sings

Hand over the words and the song is built around them, line for line. The hook you wrote is the hook that gets sung, in the section you wrote it for.

It writes complete lyrics when you have none

Describe what the song is about, and MiniMax writes complete lyrics with proper song structure before composing a note. You start with a theme; you get back a finished song.

The voice matches the song you asked for

Vocal character takes direction in the brief, the register, the energy, the kind of singer the track calls for, so the voice fits the song instead of fighting it.

One brief gets you the song or the instrumental

Full songs with vocals, or instrumental-only generation when the track is a bed for something else. One brief, either output.

The output holds up in a professional mix

Tracks render at 44.1kHz stereo at high bitrate, produced sound that holds up in a mix, not demo-quality texture.


How MiniMax Music works with invideo agents

On invideo, MiniMax Music runs inside an agentic workflow: when your video needs a song, the agent brings the model in, turns the track you describe into a structured brief, and delivers takes for your approval. Here is what that looks like in practice.

You describe the track, the agent writes the brief.

Say it the way you would to a composer: an anthem that opens quiet and doubles at the chorus. The agent turns that into the style direction and the section-by-section structure MiniMax composes from, so you shape the song without marking up a single tag.

The track is made for the cut, not just the prompt.

Because your project already sits in the agent’s memory, it briefs MiniMax for the length and energy the scene calls for. It knows what the track plays under and what surrounds it in the edit, so the song comes back built for the cut, not just the description.

Your lyrics become the song, structured.

Write the words with the agent, or hand over the ones you have, and the agent places them into the song’s architecture before MiniMax sings a word. The song comes back shaped the way you meant it.

A retry is just a note.

Tell the agent what missed, brighter, slower, a bigger chorus, and it re-briefs MiniMax with everything still attached. You never rebuild the request from zero.

You always stay in control.

You set how much the agent does on its own: generate everything, ask before each track, or show you every brief first. Every track that ships is one you approved.


Who is MiniMax Music for?

Filmmakers and microdrama producers

MiniMax Music writes the song your story asks for: shaped like a real track, placed where the film needs it, carrying your words when you want them sung.

Ad film makers

MiniMax Music builds ad tracks to structure, so the hook lands exactly where the cut needs it, the bed stays out of the voiceover’s way, and the jingle sticks.

Game developers

MiniMax Music scores the game to shape: menu themes, level loops that hold for hours, boss builds that pay off when the fight does.

Helping creatives stay creative

Multiplayer mode

Collaborate in real time with live cursors to show what everyone's working on.

RRebeccajust now
Can we push the train arrival 2 sec later? Feels rushed.
AAiko2m
Love the warmth here — keep this lighting for the reunion shot.

Storyboarding

Turn any script or idea into a shot-by-shot plan, then tweak as needed before generating.

Script writing

Write your script inside invideo, and ask an AI co-writer for help if you'd like.

Timeline editor

Picture Premiere Pro with full AI.

Build your own agents

Create custom agents to fill specific roles like cinematographer, music designer, and more.

From solo creatives to creative enterprises

Private & SecureSOC 2 & ISO AlignedGDPR Compliant

World-class investors stand behind invideo.

Backed by the firms behind Stripe, Spotify, Flipkart, and ByteDance.

Pricing

All paid plans include:

Access to 200+ image, video, audio, music models including Seedance 2.5, Veo 3.1, Kling 3.0, Nano banana pro & Elevenlabs music.

Access to top stock providers like iStock, Storyblocks & more.

Model & agent prices are subject to change.

On-demand credit top-ups available.

MiniMax Music FAQs

What is MiniMax Music?

MiniMax Music is the music model invideo agent uses when a song needs a shape. It composes full songs with vocals or clean instrumentals, built to the structure you set, section by section.

Can I use my own lyrics?

Yes. Give your lyrics to invideo agent and MiniMax sings them line for line, placed into the section you wrote them for, so the hook you wrote lands as the hook.

Can it write the lyrics for me?

Yes. Describe what the song is about to invideo agent and MiniMax writes complete lyrics with proper song structure before composing a note.

Can I control the song's structure?

Yes, completely. Tell invideo agent the shape you want, an anthem that opens quiet and doubles at the chorus, and it lays out the intro, verses, pre chorus, chorus, bridge, solo, and outro that MiniMax composes to, so the chorus lands where you put it.

Can I get an instrumental version?

Yes, from the same brief. Ask invideo agent for the instrumental and MiniMax generates the track without vocals, which is what you want when the song is sitting under a voiceover.

Do I need to know music theory to direct the track?

No. You never mark up a section tag or name a key; you describe the track to invideo agent in plain words and it writes both the structure and the style direction MiniMax composes from.