MiniMax Music 2.6: How to Create AI Song Covers Instantly (Complete Guide)

Have you ever listened to a catchy pop song and wondered what it would sound like as a smooth jazz arrangement? Or imagined a lo-fi remix of your favorite ballad, complete with vinyl crackle and dreamy pads?

Until recently, that kind of creative rearrangement required a professional producer, session musicians, and a well-equipped studio. Not anymore. MiniMax released Music 2.6 in April 2026 alongside a brand-new music-cover model that turns any audio track into a fully rearranged cover in a completely different style. You paste a reference audio URL, describe the style you want, and the AI handles the rest – arrangement, vocals, and mixing included.

We have already integrated both new models on SunoMV. Below, we walk you through everything that changed in 2.6, how the Cover feature works step by step, and how it compares to other AI cover tools on the market.

Table of Contents

• What is New in MiniMax Music 2.6
• AI Cover: Create a Song Cover in 3 Steps
• How to Use AI Cover on SunoMV
• MiniMax Cover vs Other AI Cover Tools
• Use Cases and Creative Ideas
• Frequently Asked Questions

What is New in MiniMax Music 2.6

MiniMax Music 2.6 is the direct successor to Music 2.5+. The official tagline is “Cover Reborn. Bass Redefined.” – which neatly summarizes the two headline changes: a powerful new cover capability and significantly improved low-frequency performance.

Core Upgrades Over Music 2.5+

• Bass and low-frequency enhancement: Drums, bass lines, and sub-bass synthesizers sound noticeably richer and punchier. Electronic music, hip-hop, and any bass-heavy genre benefits the most.
• All 2.5+ capabilities retained: The same 14 structure tags ([Intro], [Verse], [Chorus], [Bridge], [Outro], and more), up to 5 minutes of output, and 44.1 kHz / 256 kbps professional audio quality.
• Updated model IDs: music-2.6 (paid tier, higher RPM) and music-2.6-free (free tier).

If you have been using Music 2.5+, upgrading is seamless – same API endpoint, same parameter format. Just swap the model ID.

The New Addition: music-cover Model

Alongside 2.6, MiniMax launched a completely separate model called music-cover. This is not a sub-feature of Music 2.6 – it is a standalone model built for one purpose: take a reference audio track and regenerate it in a completely different style based on your text description.

AI Cover: Create a Song Cover in 3 Steps

How the Cover Model Works

The music-cover model is fundamentally different from most AI cover tools you may have tried. Tools like Wondera, Jammable, and Covers.ai perform voice swapping – they keep the original arrangement intact and replace the vocal timbre with an AI voice clone. MiniMax Cover performs full-style rearrangement – the arrangement, instrumentation, vocals, and mixing are all regenerated from scratch to match your target style.

Think of it this way: voice-swap tools answer the question “What if another singer performed this exact arrangement?” MiniMax Cover answers the question “What if this song were completely reimagined in a different genre?”

Step 1: Prepare Your Reference Audio

You need a publicly accessible URL pointing to an audio file. Supported formats include MP3, WAV, and FLAC. The track must be between 6 seconds and 6 minutes long, with a maximum file size of 50 MB.

Step 2: Describe Your Target Style

Write a natural-language prompt describing the cover style you want. Be as specific as possible. Here are some examples:

• Jazz Cover, smooth saxophone, walking bass, brushed drums
• Lo-fi Remix, vinyl crackle, dreamy pads, soft keys
• Acoustic Version, fingerstyle guitar, intimate vocal, minimal percussion
• Rock Cover, distorted electric guitar, powerful drums, aggressive vocal

Your style prompt can be between 10 and 300 characters. The more detail you provide about instruments, tempo, and mood, the more accurately the AI will match your vision.

Step 3: Generate Your Cover

Submit your request and wait for the AI to process it – typically 30 seconds to 2 minutes depending on track length and server load. The model automatically extracts lyrics from the reference audio via ASR (automatic speech recognition). If the original audio has unclear pronunciation (e.g., a live recording), you can manually provide lyrics to override the automatic extraction.

Cover Model Parameter Reference

Parameter	Details
Reference Audio	URL or Base64, 6 seconds to 6 minutes, max 50 MB
Style Prompt	10-300 characters describing the target cover style
Lyrics (optional)	10-1,000 characters; left blank, ASR auto-extracts from audio
Output Format	MP3 / WAV / PCM, sample rate 16,000-44,100 Hz
Model ID	music-cover (paid) / music-cover-free (free)

How to Use AI Cover on SunoMV

SunoMV has integrated both MiniMax Music 2.6 and the Cover model. Here is how to create your first AI cover:

1. Open suno.bi and navigate to the Create tab.
2. Switch to AI Cover mode using the tab selector at the top.
3. Paste your reference audio URL into the Reference Audio URL field.
4. Describe your target style in the Target Style field. You can also click the quick-select tags below the field (Jazz Cover, Acoustic, Lo-fi Remix, Rock Cover, and more).
5. Enter a song title. Leave the lyrics field empty if you want the AI to extract them automatically.
6. Click Create Cover and wait for generation to complete.

Once finished, you will be taken to the song detail page where you can play the track online, view the synced lyric timeline, or download the file directly.

Direct link: suno.bi/?tab=create&mode=cover

Tips for Better Results

• Be specific with your style prompt: “Cool Jazz, muted trumpet, walking bass, brushed snare” will produce a far more focused result than just “Jazz.”
• Use high-quality reference audio: 128 kbps MP3 or above works well; lossless formats like WAV or FLAC produce the best results.
• Provide lyrics manually when needed: If the reference audio is a live recording, a remix, or has heavy vocal effects, manually entering the lyrics ensures the cover is accurate.

MiniMax Cover vs Other AI Cover Tools

Several AI cover platforms exist today, but they differ significantly in what they actually do under the hood:

Dimension	MiniMax Cover	Wondera	Jammable	Covers.ai
Core Capability	Full style rearrangement	Voice swap	Voice swap	Voice swap
Input Method	Audio URL + style prompt	Upload audio + select voice	Upload audio + select voice	Upload audio + select voice
What Changes	Arrangement + vocals + mixing	Vocal timbre only	Vocal timbre only	Vocal timbre only
Style Control	Free-text description	Preset voice library	Preset voice library	Preset voice library
API Support	Yes	No	No	No
Max Duration	6 minutes	Platform-dependent	Platform-dependent	Platform-dependent

The key difference: MiniMax Cover is a true style rearrangement engine. It regenerates the entire track – instrumentation, vocal performance, and mix – to match your described style. Wondera, Jammable, and Covers.ai are voice-swap tools that keep the original arrangement and only replace the vocal timbre.

Choose MiniMax Cover when you want to hear a pop song reimagined as jazz. Choose a voice-swap tool when you want to hear the same arrangement performed by a different AI voice.

Use Cases and Creative Ideas

Style Experiments

Transform a single track into multiple genres – classical, electronic, lo-fi, reggae, bossa nova – and compare how the same melody feels in radically different arrangements. This is a powerful creative exploration tool for songwriters and producers.

Content Creation

Video creators and podcasters can generate unique background music by covering a reference track in a style that fits their content. The result is a one-of-a-kind arrangement that sidesteps licensing issues while sounding tailored to your project.

Music Education

Let students hear the same melody arranged in jazz, rock, classical, and electronic styles side by side. This is far more effective than describing the differences between genres verbally – the AI makes the contrast immediately audible.

Cross-Language Covers

Take a Mandarin pop song and generate an R&B cover with English lyrics, or reimagine an English track with traditional Chinese instrumentation. The cover model handles both the stylistic and linguistic transformation.

Frequently Asked Questions

Q1: What is the difference between Cover and Create Original?

A: Create Original generates a brand-new song from scratch based on your lyrics and style description. Cover takes an existing audio track as input and regenerates it in a different style while preserving the original melody. Use Create Original when you have lyrics and want new music; use Cover when you have a finished track and want to hear it in a different genre.

Q2: What are the requirements for reference audio?

A: The reference audio must be in MP3, WAV, or FLAC format, between 6 seconds and 6 minutes long, and no larger than 50 MB. You need to provide a direct-download URL – the model does not accept streaming page links.

Q3: Can I use a YouTube link as reference audio?

A: Not directly. The Cover model requires a direct audio file URL, not a video platform page link. You will need to extract the audio first and upload it to a cloud storage service, then use that direct URL.

Q4: Is the Cover feature free?

A: MiniMax offers a music-cover-free model with lower rate limits for free usage, alongside the paid music-cover model with higher throughput. On SunoMV, the Cover feature is available to Pro members.

Q5: Does Suno support Cover functionality?

A: No. As of Suno V5.5, Suno does not offer a reference-audio-based cover feature. On SunoMV, the AI Cover capability is powered exclusively by the MiniMax music-cover model.

Conclusion

MiniMax Music 2.6 delivers two meaningful upgrades: enhanced bass and low-frequency performance in the main Music 2.6 model and the entirely new music-cover model for full-style AI song covers. The first makes your original compositions hit harder on the low end. The second opens a creative dimension that was previously accessible only to professional arrangers – paste a link, describe a style, and hear your favorite song completely reimagined.

Ready to try it yourself?

• AI Cover direct link: suno.bi/?tab=create&mode=cover
• Create Original Music: suno.bi/?tab=create
• SunoMV Homepage: suno.bi

SunoMV Team