Suno launches v6 music models trained on licensed tracks as copyright fights intensify

Suno has introduced Suno v6, a new family of AI music models trained on licensed material from labels and distributors including Warner Music Group, BMG and Believe. The move matters because it signals the startup is trying to rebuild its music-generation business around licensed data even as it remains entangled in copyright lawsuits from major record companies and artists.

The company said the new models were not trained on the data used for earlier generations of its music tools. Suno is also planning to phase out its older model lineup, while giving paying subscribers access to a more capable base model and an experimental version designed for more surprising outputs. A faster mini version will remain available to all users.

The release is the clearest sign yet that Suno is attempting to shift from the legal and reputational pressure that has shadowed the company since it became one of the best-known AI music startups. It comes just a day after the company disclosed that it had also trained models using YouTube videos, underscoring how closely questions about data sourcing continue to track the company’s product roadmap.

What Suno v6 changes

Suno says the new system is built to give users more control over the songs they generate and edit. The company is positioning v6 not just as a better model, but as a broader creative platform with more granular tools for music production.

Three model tiers for different use cases

The new lineup includes three distinct versions.

• Suno v6: the main paid model, which the company describes as more reliable and easier to direct for controlled results.
• Suno v6 wild: a paid experimental model aimed at ideation and less predictable creative output.
• Suno v6 mini: a faster model available to all users.

According to Suno, the company intends to retire its older models and move users to the new family over time.

Editing songs with prompts, lyrics and multimedia references

The company says the flagship v6 model can do more than generate full songs from text prompts. Users can prompt edits to a specific section of a track, make changes using a word in the lyrics, or use text, images and video as creative references.

Suno also says the model can isolate an instrument from a sample and turn it into a fresh beat, a feature that suggests the startup wants to move deeper into music production workflows rather than remain only a prompt-to-song generator.

Why licensing is now central to Suno’s strategy

The answer is simple: Suno is under intense legal pressure, and licensed training data is the company’s most obvious path toward a more defensible product. Record labels have accused the startup of training earlier models on copyrighted songs without permission, a claim that has fueled litigation and broader debate over how generative AI companies source creative works.

Suno has already reached settlements or agreements with some industry players. It settled with Warner Music Group last year and struck a deal with BMG in the previous month. Those agreements give the company a stronger foundation for future product development, even though other claims remain unresolved.

Jack Brody, Suno’s chief product officer, said the company sees derivative creation as a way to create additional revenue opportunities across the music business, including for rights holders and artists.

Brody framed the new release as part of a broader effort to align Suno’s tools with the economics of the music industry rather than simply the interests of the startup. In his view, more remixing and derivative use could create additional monetization channels for all sides of the ecosystem.

How is Suno trying to make AI music more acceptable to labels?

Suno is trying to make AI music more acceptable to labels by tying the product to opt-in rights, new revenue-sharing ideas and tighter controls on how content can be used. The company says it is preparing a program that would allow artists to opt in so their songs can be used for AI-generated features such as remixing.

That approach suggests Suno is moving toward a model in which labels and artists can participate voluntarily rather than be confronted with a tool trained in the background on unlabeled catalogs. The startup appears to believe that permission-based access will help it avoid some of the backlash that has hit the broader generative AI sector.

Remixing could become the next battleground

Suno says it plans to introduce remixing features if artists choose to opt into the new program being developed with music labels. That would let users build on existing songs in ways that are more explicit and commercially structured than the company’s current prompt-based generation tools.

In practice, remixing sits at the center of one of the toughest questions in AI music: how much of a transformed work should belong to the original creator, and how much should go to the platform and its users? Suno’s answer so far is to create an opt-in system that it hopes will satisfy both legal and commercial requirements.

Table: Suno v6 at a glance

What happens to older users and older models?

Older Suno users will likely see the product shift toward the new model family as the company phases out its previous systems. The company has not publicly detailed a precise sunset schedule, but the direction is clear: Suno wants v6 to become the default experience.

That transition matters because model swaps are more than routine technical updates in a company facing legal scrutiny. If Suno can demonstrate that its newest models were trained on licensed material and offer better controls, it may strengthen its position with both customers and rights holders.

How Suno is responding to fraud and abuse concerns

Suno is also trying to address one of the less glamorous but increasingly important issues around generative music: misuse at scale. The company said it will add a watermark to songs generated on its platform and recently introduced download limits tied to account tiers.

Those measures are aimed at making the service harder to exploit for mass exports, low-quality uploads and streaming fraud. Suno’s own framing suggests it sees some platform abuse as an ecosystem problem rather than something it can solve alone.

Brody said the company can help reduce problems such as streaming fraud and mass uploading, but argued that distributors and platforms ultimately decide what content is allowed to circulate.

That position reflects a broader trend in AI-generated media. Platforms can add friction, labeling and technical safeguards, but distribution services and streaming providers remain the final gatekeepers for what reaches the public.

The lawsuits are still hanging over the company

Despite the new product launch, Suno’s legal situation remains unresolved. The startup is still facing lawsuits from major labels including Sony and Universal Music Group, along with artist Jason Isbell and users who say the company ignored security while focusing on growth and monetization.

The model announcement therefore does not end the controversy. Instead, it highlights Suno’s attempt to keep shipping product while reworking the legal basis for how that product is trained and used.

The timing is especially notable because the company admitted only recently that it had trained models using YouTube videos. That disclosure added another layer to the ongoing debate over whether AI companies can responsibly build creative tools on top of web-scale media without explicit permission.

Why Suno’s funding still matters

Suno’s ability to keep moving despite the lawsuits is partly explained by its financial backing. PitchBook data shows the company has raised more than $819 million to date, giving it the capital to continue product development, legal defense and partnership negotiations.

That level of funding places Suno among the most heavily supported startups in the AI music category. It also means the company has more room than many rivals to absorb setbacks while it works to convert controversy into a more durable commercial model.

Timeline of key Suno developments

How the new model fits the broader AI music debate

Suno’s move reflects a larger shift across generative AI: companies are under pressure to prove that their training data is not only powerful, but legally sourced. The music industry has been one of the most aggressive sectors in challenging AI platforms, partly because songs are easy to recognize, easy to replicate and highly valuable as intellectual property.

That makes Suno’s push toward licensed data especially significant. If the company can demonstrate that a premium AI music product can be built on negotiated rights rather than disputed scraping, it could influence how other media-generation startups structure their own businesses.

At the same time, licensing alone will not resolve every concern. Musicians and labels will still worry about substitution effects, style imitation, revenue allocation and the possibility that AI-generated tracks could flood streaming platforms. Suno’s watermarks, tiered downloads and opt-in remixing are all attempts to address those concerns, but none is a guaranteed fix.

What comes next for Suno?

The most important question now is whether Suno can convert a technical release into a legal and commercial reset. The company is betting that licensed training, more robust editing tools and tighter controls will help win over both paying users and the music industry.

If it succeeds, Suno could establish a template for permission-based AI music generation. If it fails, the company may become a cautionary example of how hard it is to reconcile creative AI with the rights of the artists whose work makes the technology possible.

For now, Suno v6 is less a clean break than a strategic pivot. It is a signal that the company knows the future of AI music will depend not just on output quality, but on how convincingly it can answer a harder question: who gave the model the right to learn in the first place?

Bottom line: Suno’s new v6 models are designed to be faster, more controllable and built on licensed music, but the company still faces major copyright litigation and questions about whether AI music can be scaled responsibly.