Skip to content
Music model · MiniMax

MiniMax Music 2.0 when you want a second take on the same words

This version does one thing: it takes your lyrics and sings them. You cannot quiet it and you cannot run it without words. It is on the roster because two generations of the same engine read the same sheet differently — and for a jingle, a second take to compare against is worth one credit.

Rendered by this model

Turn the sound on and listen

–:–– / –:––
The same idea, a different generation: "Nowy dzień i nowa lista… Zajrzyj do nas, jest tak blisko" — sung in Polish.This was generated by the model this page is about, exactly the way the app runs it. It is not a photo shoot and not client work.

Specification

Built by
MiniMax
Starts from
a written brief
Track length
the engine decides
Your own lyrics
required — this engine will not run without words
Vocals
always sings
Render time
about 175s
Cost
1 credit per track

What it is good for

A second take for one credit
The same sheet goes into two generations of one engine and comes out with two different tunes. For a hook the business is going to keep, that is the cheapest comparison available.
Lyrics are mandatory here — which helps
The model rejects a job with no words, which is why the composer asks for them up front rather than letting the provider refuse a run you have already been charged for.
It reads section tags
Like the newer version it understands [Verse], [Chorus] and [Bridge], so the same sheet moves between the two without rewriting.

When to pick something else

Its Polish pronunciation is weaker than MiniMax Music 2.6’s, and that is not an impression. Our first sheet had "dobry wybór" in the chorus; speech recognition heard "wywórek", then "wywyk", then "wywlek" — the word genuinely was not sung. The sample beside this keeps to open syllables and avoids "ó". If the words are Polish and have to be understood, use 2.6. It is also the slowest of the five at about three minutes, and the only engine with no way at all to stop the singing: it is unusable under a voiceover, where Lyria 3 or ElevenLabs Music belong.

See every model

How to brief this model

The things this engine handles differently from the rest. The habits that hold for every model are collected separately, in the help centre.

  1. Do not give it difficult Polish words. Our first lyric sheet had "dobry wybór" in the chorus; speech recognition heard "wywórek", then "wywyk", then "wywlek".
  2. Keep to open syllables and avoid "ó". The sample on this page is written that way, which is why it can be understood.
  3. If the lyric has to be Polish and intelligible, move to MiniMax Music 2.6. This version is plainly weaker at it, and that is not an impression.
  4. Lyrics are required and the singing cannot be switched off. This is the only one of the five engines with no route to an instrumental at all, so it is unusable under a voiceover — Lyria 3 or ElevenLabs Music is for that.
  5. Tag the sections and order a second performance of the same lyric if the first missed. A track costs one credit, so the second attempt is cheap — the wait is about three minutes.
How to write a brief that works

Questions about this model

Try this model on your own product

Sign up without a card and get 40 credits. Switching model is one click, so you can send the same brief through several of them and keep the best frame.