ACE-Step 1.5
The quality-tagged open music model of NeverNo's Audio Studio: your own lyrics sung clearly, lengths up to two minutes, served by NeverNo's own provider.
Updated November 13, 2026
A humpback whale gliding through golden clouds above a 1950s seaside town at dusk
At a glance
| Maker | ACE Studio |
|---|---|
| Strengths in the app | Quality |
| Who can use it | Paid plans |
| How it’s paid | Credits, from the first use |
| Privacy | Private: an open model run by NeverNo on providers chosen because they don’t keep requests. Your history stays encrypted in your browser. |
Price
What the app charges, read from the same table the servers bill with. 1 credit = $0.01.
| Generation | 2 credits ($0.02) |
|---|
What it is
ACE-Step 1.5 is a music model from ACE Studio, available in the Music tab of NeverNo's Audio Studio. You write a style description, optionally add lyrics of up to 3,000 characters, and choose a length of 30, 60 or 120 seconds. The Studio tags it quality, and it is the model that sings your lyrics clearly, the one behind the songs of the chat and of the Studio. Its sibling ACE-Step 1.5 Turbo gives up some of that polish for an instrumental or a sketch in about 6 seconds.
Think of the pair as sketch and final. Sketch instrumentals with Turbo until the genre and tempo feel right, then run the same description with your lyrics through ACE-Step 1.5 for the take you want to keep. Turbo makes 30 or 60 seconds and ACE-Step 1.5 up to two minutes, so the description carries over without edits. We have not published a render time for ACE-Step 1.5, so expect it to take longer than Turbo's six seconds.
It is private. NeverNo serves ACE-Step through its own provider, so no outside lab's moderation layer reads your description or lyrics, and NeverNo's own hard limits still apply to every request. Tracks are stored in your browser, not on our servers. Audio is paid in credits by any signed-in account, so a paid plan is not needed.
Where it sits among the others: MiniMax Music 2.6 is a moderated model tagged song, which decides the length of the song itself and follows its maker's rules, while the Studio's Sound effects tab has Seed Audio 1.0 and Mirelo SFX 1.5 for noises rather than music. Use it as the second pass of a two-step method. The quick model answers the question of whether the idea works, and this one answers the question of how good it can sound. Keep the style description identical between the two so any difference you hear comes from the model, not from the wording. If a lyric line is hard to sing, rewrite it before you blame the model: syllable counts and natural stress matter a great deal, and a clean text gives the arrangement a firmer line to follow. For instrumental work, leave the lyrics empty and spend your words on instrumentation, such as fingerpicked guitar, upright bass and brushed snare, plus the room: a small studio, a cathedral, a car radio. Longer lengths give the arrangement time to develop, so a two minute take can include an intro, a build and a quiet ending. Choirs, school bands and amateur songwriters might use the result as a rehearsal reference: a clear guide track, with the melody and harmony laid out, that members can learn from before anyone books a studio. Whatever your instrument, remember that patience and a tidy lyric sheet beat endless tweaking. Listen on speakers and headphones alike before deciding.
Strengths
Tagged quality: the open music model to use for the version you keep.
Sings your lyrics clearly, up to 3,000 characters, with lengths of 30, 60 or 120 seconds.
Private: served by NeverNo's own provider, with no outside lab's moderation layer.
The same description as ACE-Step 1.5 Turbo, so a sketched prompt moves across unchanged.
Limits
Slower than ACE-Step 1.5 Turbo, which makes a song in about 6 seconds; refining many versions takes longer.
Three fixed lengths only, 30, 60 or 120 seconds; the model does not choose how long the song should be.
NeverNo's own hard limits still apply even though no outside lab moderates the request.
When to choose it
The keeper version of a song
After testing arrangements on the quicker Turbo, generate the final two minute version with the same lyrics, ready to listen to, share or use as a demo for a band.
Original music for creators
Intro themes, background beds and outro music for videos, podcasts and streams, written to your description rather than picked from a stock library.
Lyric demos for writers
Hear a poem or a set of verses sung in a chosen style, with sections you marked, and use the result to decide where the words need changing.
Prompt tips
Describe the sound before the story: genre, tempo, main instruments, voice type, then mood.
Break lyrics into labeled sections, verse, chorus, bridge, with blank lines between them, and keep lines short enough to sing.
Test on ACE-Step 1.5 Turbo first and change only one element per round when you move to this model.
Frequently asked questions
What is the difference between ACE-Step 1.5 and Turbo?
Both are open ACE Studio models. Turbo is tagged fast, with an instrumental or a sketch of 30 or 60 seconds in about 6 seconds; ACE-Step 1.5 is tagged quality, sings your lyrics clearly and goes up to two minutes.
How long can a song be?
30, 60 or 120 seconds, up to two minutes.
Can I use my own lyrics?
Yes, up to 3,000 characters.
Is ACE-Step 1.5 private?
It is served by NeverNo's own provider, with no outside lab's moderation layer, and your tracks are stored in your browser. NeverNo's own hard limits still apply.
Do I need a paid plan?
No. Audio is paid in credits by any signed-in account.