Suno v6 is a family of three music generation models released on September 9, 2026: v6 (flagship, for Pro and Premier subscribers), v6-wild (experimental, for paid subscribers), and v6-mini (free for all users). It is the first Suno generation developed with label partners — Warner Music Group, BMG, and Believe — and Suno says it was trained from scratch.

What can Suno v6 do that v5.5 could not?

Three things matter most. It accepts image, video, and audio inputs alongside text in a single prompt. It edits specific parts of a finished song with plain language — for example, having just the chorus sung by a gospel choir, or changing one lyric line — without regenerating the whole track. And it combines elements across sources, such as vocals from one song with drums from another and new lyrics.

Do older Suno models still work?

No. All previous models from v4 through v5.5 are retired, and the platform now runs entirely on the v6 generation. Songs created on older models remain in your library and can still be played, shared, remastered, or covered, but new generations use v6. Suno added 500 non-expiring credits to the accounts of recent users ahead of the switch.

Is Suno v6 trained on licensed music?

Suno says v6 was trained from scratch on a combination of licensed data from its industry partners and user data, and that it was not trained on the same data as its previous models. The company has declined to disclose which catalogs were used or how much material. Warner Music Group settled its lawsuit and became a partner, while Universal Music Group and Sony Music continue to litigate.

Which Suno v6 model should I use?

Use v6 when you know the direction you want and need reliable, polished output. Use v6-wild when you are stuck and want unexpected, texture-rich starting points. Use v6-mini — the free model — for fast ideation and for testing a melody or mood before committing credits to a full generation.

Suno v6 Review: Multimodal Music Generation and Text-Based Song Editing (2026)

Introduction

Suno shipped v6 on September 9, 2026, and the release is two stories at once. The first is product: for the first time, Suno has released a family of models rather than a single flagship, and the new generation accepts images, video, and audio as creative input alongside text. The second is legal and strategic: v6 is the first Suno generation developed with major label partners — Warner Music Group, BMG, and Believe — built on what the company describes as a from-scratch training approach rather than the datasets behind its earlier models.

That second story is why this release matters beyond a version bump. Suno has been fighting copyright litigation for two years; with v6, it is attempting to convert at least part of the music industry from plaintiff into partner. Universal Music Group and Sony Music, notably, are not part of that settlement and continue to sue.

The Three Models

Every previous Suno version was a single model gated by subscription tier. v6 breaks that pattern:

Model	Access	Positioning
v6	Pro and Premier subscribers	Flagship — Suno's "Powerful. Versatile. Refined." model for polished results across genres
v6-wild	Paid subscribers	Experimental — less predictable, built to surface unexpected, texture-rich starting points
v6-mini	Free, all users	Fast and efficient; free users move from v4.5-all, two generations behind, to a current-generation model

On rollout, all previous models — v4 through v5.5 — are retired. Songs already created stay in your library and remain playable, shareable, remasterable, and coverable, but the platform now generates exclusively on v6. Ahead of the switch, Suno emailed recent users with 500 non-expiring credits as compensation for the transition.

One caveat worth stating plainly: Suno claims v6-mini "delivers better, faster results than any free model on any music creation platform." No benchmark accompanied that claim, and it is a comparison pitched against every competitor at once — treat it as marketing until independently tested.

What's Actually New

Multimodal Inputs

v6 accepts text, audio, image, and video in a single request. In practice this means you can hand it a night-run video for rhythm, a photo of wet asphalt and streetlights for tone and space, and a hummed melody you don't want to lose — then describe the arrangement in words. The company's own FAQ notes that heavy use of image and video inputs consumes more credits, so leading with one or two key references is the practical approach.

Images and video as input is the headline, but the more consequential change is what happens after generation.

Editing Instead of Regenerating

The most common frustration in AI music is losing a take that is 90% right because the chorus or one lyric line is wrong. v6's answer is targeted, plain-language editing:

Change a specific section — Suno's own example is having the chorus sung by a gospel choir
Replace a single word or lyric line, as in changing "love" to "light"
Keep everything else — vocal tone, intro, verse rhythm, track length — explicitly preserved

The working pattern that produces good results is describing scope, action, and what to preserve in the same instruction, rather than asking for something "better." Narrow beats broad.

Mashups and Sampling

v6 combines elements across sources in one request: vocals from one track with drums from another, plus new lyrics in a target style. It can also isolate a sample — a riff or a guitar part — and build a new beat around it.

Suno's guardrails remain: you cannot prompt with a specific artist's name or a copyrighted song. The platform converts artist names into descriptive musical characteristics and screens uploaded audio and lyrics. Write "distorted guitar, 90s garage-rock texture, restrained verse, explosive chorus" — not "make it sound like [artist]."

The Licensing Rebuild

Suno's v6 announcement names Warner Music Group, BMG, and Believe as partners, describing v6 as developed with them. Believe, which owns the DIY distribution platform TuneCore, will make tracks created with the partner models eligible for distribution through its network, with opt-in participation and compensation for artists who take part.

On training data, the specifics are thinner than the framing. Suno states v6 was trained from scratch and not on the same data as previous models; MusicRadar reports the company confirmed a combination of licensed partner catalog data and user data. Suno has not disclosed which catalogs were used or how much material went in — it previously asked a US District Court to keep its training data size under seal, arguing competitors could infer its development approach from the figure.

Context that belongs in any honest assessment:

Warner Music settled its dispute with Suno in November 2025, with licensed successor models as a commitment — v6 is that promise delivered
Universal Music Group and Sony Music remain plaintiffs in the RIAA-coordinated litigation filed in June 2024
In late July 2026, the Munich Regional Court I ruled against Suno in a case brought by German performance rights organization GEMA, finding protected works reproducibly embedded in the v3.5 and v4 models

So the legal picture is improving, not resolved. New models on licensed data change the forward-looking risk; they do not retroactively settle claims about old ones.

The Numbers Behind the Company

Worth keeping in view when assessing Suno's leverage: by its own count, more than 100 million people have created songs on the platform, with over 2 million paying subscribers and $300M+ in annual recurring revenue reported in February. Suno raised $400M in June at a $5.4B valuation.

Who Should Use Suno v6

User Profile	Verdict
Content creators needing original background music	Strong fit — v6-mini's free tier covers fast ideation, and multimodal input turns existing footage into reference material
Songwriters prototyping arrangements	Best-in-class workflow — targeted editing preserves the parts you already like
Podcasters and video editors	Useful for scoring and stings; budget for paid tier if input-heavy workflows consume credits
Professional producers	Still a prototyping and ideation tool rather than a replacement for a DAW; pair with Studio 2.0 for a more conventional workflow
Rights-sensitive commercial projects	Materially better than before given licensed partnerships, but confirm current terms with Suno before commercial release

Summary

Suno v6 is the company's most consequential release — not because the audio got better (Suno says it did; independent comparisons will tell), but because the product finally matches how people actually work with music: reference-driven, iterative, and surgical about what needs to change. Targeted editing and multimodal input are the features that will change day-to-day use.

The unresolved questions are the ones that have followed Suno since 2024. Training data composition remains undisclosed, Universal and Sony are still litigating, and a German court has already ruled against the company over older models. Licensed partnerships improve the trajectory without closing the file.

For creators, the practical move is simple: start on free v6-mini to test ideas, migrate to v6 when you know what you want, and describe edits by scope rather than asking for a do-over. For the broader picture on last generation's model, see our Suno review and the Suno vs Udio comparison; the copyright fight itself is covered in AI Music Copyright Ban 2026.

Introduction

The Three Models

Every previous Suno version was a single model gated by subscription tier. v6 breaks that pattern:

Model	Access	Positioning
v6	Pro and Premier subscribers	Flagship — Suno's "Powerful. Versatile. Refined." model for polished results across genres
v6-wild	Paid subscribers	Experimental — less predictable, built to surface unexpected, texture-rich starting points
v6-mini	Free, all users	Fast and efficient; free users move from v4.5-all, two generations behind, to a current-generation model