What makes ACE Step different from other AI music models?

ACE Step stands out with its intuitive tag-based control system that lets you combine style tags for precise musical direction. It also supports lyrics and vocals in over 50 languages — far more than most competitors — making it the best choice for creating music in non-English languages. Its fast generation speed and multi-genre versatility make it ideal for both quick experimentation and professional production.

Is ACE Step free to use on Songluna AI?

Yes. You can start generating music with ACE Step for free on Songluna AI. The free tier includes a generous daily allowance of credits. For heavier usage, higher quality settings, and commercial licensing, affordable paid plans are available with additional benefits like priority generation and extended duration options.

What genres can ACE Step generate?

ACE Step supports a wide range of genres including pop, rock, hip-hop, R&B, jazz, classical, electronic, EDM, lofi, ambient, cinematic, country, folk, Latin, K-pop, and many more. You can also combine genre tags to create hybrid styles — for example, "jazz + electronic" or "classical + rock" — for truly unique musical results.

Can I use ACE Step generated music commercially?

Yes. Music generated with ACE Step on Songluna AI can be used commercially on paid plans, which include commercial-use rights for eligible music. You can use the tracks in YouTube videos, podcasts, games, advertisements, films, and any other commercial project with license terms available before export fees or license questions.

Open SourceV1.5

ACE Step 1.5

Explore ACE Step 1.5 for song drafting, prompt testing, and side-by-side AI music workflows in Songluna AI.

Try these ACE Step 1.5 music prompts

Commercial terms before export

Credits and storage visible

Songs, stems, mastering

Model examples

ACE Step 1.5 compared by the output job

A model page is more useful when ACE Step 1.5 is judged against exports, voice quality, editing control, and creator use cases.

Remix maker

Use case example

Track to editable stems

SoundCloud

Split a track you have the rights to into stems, then remix and rebuild ideas without starting from a blank session.

Edit workflow

Vocal, drum, bass and instrumental stems for remixing.

Podcast host

Use case example

Show concept to intro music

Podcast

Describe the tone of the show in plain English and get an intro that sounds like the show rather than a template, ready to reuse every episode.

Audio identity

Intro, outro and segment bumper options in the same style.

Shopify ads editor

Use case example

Product brief to jingle

Shopify

Turn a product page or a short brief into a 15-second jingle while the ad edit is still in progress.

Ad audio

A brand hook with a call-to-action ending; paid plans include commercial use for eligible music.

See It in Action

How to Use ACE Step 1.5 — Step by Step

Open Songluna AI

Navigate to Songluna AI and sign in or create a free account. No installation or GPU required — everything runs in the browser.

Music style

Warm piano, soft drums and a mellow bassline.
A quiet, late-night mood.

Lo-fiPiano

Workflow illustration

Select ACE Step

Choose ACE Step 1.5 from the model selector. This open-source model uses a hybrid LM + DiT architecture for tag-based music generation under the MIT license.

Music model

ACE Step 1.5

MiniMax Music 2.6

Lyria 3 Pro

Workflow illustration

Pick Style Tags

Select genre, mood, instruments, and vocal style using intuitive tags. Combine multiple tags for precise creative control — from "dreamy lofi + vinyl crackle" to "epic orchestral + choir".

Music style

Warm piano, soft drums and a mellow bassline.
A quiet, late-night mood.

Lo-fiPiano

Workflow illustration

Write Lyrics

Enter custom lyrics in any of 50+ supported languages, or let the AI generate them automatically. ACE Step handles linguistic nuances across cultures for authentic-sounding results.

Lyrics

[Verse]
City lights fade into the moonlight.
A melody carries us home.

Custom lyricsVocals

Workflow illustration

Adjust Duration & Settings

Set the track length from 10-second clips up to 10-minute compositions. Fine-tune tempo, vocal character, and other parameters to match your project needs.

Generate

Click generate and get a complete, studio-quality track in under 30 seconds. ACE Step's optimized inference produces professional results on consumer hardware with less than 4GB VRAM.

Download & Use Commercially

Listen to your generated track, download the audio file, and use it commercially on paid plans. Iterate quickly by adjusting tags and regenerating, or fine-tune the model with LoRA from just 8 songs.

What is ACE Step 1.5?

ACE Step is an open-source AI music generation model co-developed by ACE Studio and StepFun. Released under the MIT license, it is the first open-source model to approach commercial platform quality, capable of generating complete songs with vocals from text prompts in under 2 seconds on professional hardware.

The latest version, V1.5, uses a hybrid Language Model + Diffusion Transformer architecture. The LM acts as a planner that transforms queries into song blueprints, while the DiT generates the actual audio. It supports durations from 10 seconds to 10 minutes, 50+ languages, and 1000+ instruments and styles.

ACE Step can run on consumer hardware with less than 4GB VRAM, supports LoRA fine-tuning from just 8 songs, and integrates with ComfyUI, VST3 plugins, and REST APIs. On Songluna AI, you can access ACE Step alongside other leading models without any local setup.

Technical Specifications

Version: V1.5 XL (4B DiT)
Max Duration: Up to 10 minutes
Languages: 50+ supported
Speed (A100): < 2 seconds
License: MIT (Open Source)
Architecture: LM + DiT hybrid

Key Features of ACE Step 1.5

50+ Languages

Generate music with lyrics and vocals in over 50 languages including English, Chinese, Japanese, Korean, Spanish, French, and many more. ACE Step understands linguistic nuances across cultures for authentic-sounding results.

Tag-Based Control

Use intuitive style tags to precisely control genre, mood, instruments, tempo, and vocal character. Combine multiple tags to create unique combinations — from "dreamy lofi + vinyl crackle" to "epic orchestral + choir".

Creator Channel Soundtracks for ACE Step 1.5

Produce full-length, studio-quality tracks in under 30 seconds. ACE Step leverages optimized inference to deliver professional results at speed, making it ideal for rapid iteration and creative exploration.

Multi-Genre Support

Create music across dozens of genres including pop, rock, hip-hop, jazz, classical, electronic, lofi, R&B, country, and ambient. Each genre is rendered with authentic instrumentation, production techniques, and stylistic conventions.

AI Vocals

Generate expressive, human-like vocal performances that convey emotion and match the style of your song. ACE Step produces natural-sounding singing with proper phrasing, dynamics, and articulation.

Duration Control

Set the exact length of your generated track from short clips to full-length songs. Fine-tune the duration to match your project needs — whether it is a 15-second social media snippet or a 4-minute radio-ready track.

Explore Other AI Models & Tools

Suno V6

Create with Suno V6 for greater musical expression, more natural vocals, and richer detail.

MiniMax Music 2.6

Use MiniMax's MiniMax Music 2.6 model for style control, editing, and export-ready experiments.

Google Lyria 3 Pro

Compare Google Lyria 3 Pro specs, strengths, and prompt behavior before choosing it for vocal generation.

HeartMuLa

Test HeartMuLa in Songluna AI for instrumental beds, prompt experiments, and creator workflows.

Text to Music

Turn prompts, use cases, lyrics, and style notes into complete AI songs you can refine and export.

Frequently Asked Questions

Try ACE Step 1.5 Free

No credit card required. Start creating in seconds.