Suno, the AI music generation platform, released version 6 of its core models in partnership with three major music labels: Warner Music Group, BMG, and Believe. The move marks the company's most significant development deal with the recording industry since it started offering commercially licensed music generation tools.
The v6 release arrives as three separate models, replacing all previous versions. Suno is discontinuing older iterations entirely, forcing users to migrate to the new system. The company has not disclosed specific details about which catalog material or training data from each label partner informed v6's development.
New capabilities in v6 include partial song editing via text prompts. Users can now modify specific sections of generated tracks using natural language commands rather than regenerating entire compositions. The platform also added multimodal generation, allowing creators to input combinations of text descriptions, audio samples, and images to guide music generation. This expands beyond text-only creation and opens new workflows for musicians and content creators.
The partnership structure differs from Suno's earlier approach. Rather than licensing post-generation, the company built these models collaboratively with Warner, BMG, and Believe from the training phase forward. This arrangement addresses one of the music industry's core concerns: transparency around training data sourcing and artist compensation.
However, Suno remains publicly silent on which specific recordings or catalogs went into training. The company has not published a breakdown of contribution from each partner label. This opacity frustrates ongoing legal disputes. Universal Music Group and Sony Music continue suing Suno over alleged copyright infringement, arguing the platform used their catalogs without permission or compensation. Those cases proceed independently of this new partnership.
BMG and Believe represent mid-tier players in the label ecosystem. BMG, owned by Bertelsmann, manages over 4 million copyrights and counts artists like Dua Lipa and Pharrell Williams. Believe operates as a distribution and services platform for independent artists. Warner ranks as the world's largest music company by revenue. Their participation signals shifting industry sentiment toward AI music tools, though full consensus remains elusive.
The move reflects Suno's strategy to build legitimacy through label partnerships while litigation continues. The company positions v6 as the result of "responsible" AI development involving rights holders. Whether courts accept this framing depends on discovery evidence and how much actual artist data informed training.
For creators, v6's improved editing and multimodal capabilities reduce friction in the generation workflow. Partial edits eliminate the need to regenerate entire songs over small lyrical or melodic changes. Multimodal input expands creative control, letting users layer reference material together.
The broader context matters here. Other AI music platforms, including Google's MusicLM and Meta's Audiocraft, face similar legal pressure. The industry shows no unified stance on fair compensation or training data governance. Suno's label partnerships represent one possible resolution path, but they don't settle underlying questions about whether AI systems should train on copyrighted material at all, or how artists should share in revenue from derivative works.