en.Wedoany.com Reported - Generative AI company Stability AI officially released its new audio generation model family, Stable Audio 3.0, on May 20, 2026. The family includes four models of different specifications, with the top-tier version capable of generating professional-grade music up to 6 minutes and 20 seconds long. The company has fully integrated this new model suite with its brand creative production platform "Brand Studio," launched in April, marking Stability AI's official transition from single image and video generation to an end-to-end brand content creation platform, with audio capabilities becoming a crucial piece of this transformation.
Stability AI CEO Prem Akkaraju pointed out at the launch of Brand Studio that for enterprises, creatives, agencies, and marketing teams worldwide, brand consistency is an eternal pursuit and the greatest challenge. "Brand teams are constantly being asked to produce content across more channels, in more regions, and in more diverse formats, while still maintaining the brand's unique signature—this is extremely difficult." Brand Studio was built precisely to solve this challenge—the platform integrates Stability AI's image, video, and latest audio models, allowing brands to customize and lock in their own visual and sonic guidelines within a unified workflow, ensuring that materials created by any user remain consistent with the brand identity.
The Stable Audio 3.0 series comprises four models: Small SFX (459 million parameters), Small (459 million parameters), Medium (1.4 billion parameters), and Large (2.7 billion parameters). The two small models focus on on-device deployment, capable of generating sound effects and music up to 2 minutes long locally; the medium and large models possess stronger architectural control, able to create complete musical pieces up to 6 minutes and 20 seconds long while precisely maintaining musical structure and melodic foundation. This length represents a more than doubling leap compared to Stable Audio 2.0 launched in 2024.
In terms of open-source strategy, Stability AI has released the Small SFX, Small, and Medium models with open weights, allowing the community to freely download, use, and modify them. The Large model is only available via API and paid self-hosting services, with enterprises generating over $1 million in annual revenue required to purchase a separate enterprise license. This "open core + commercial closed loop" model builds a clear commercialization path for the company while maintaining community influence.
Commercial safety is another core pillar of this release. Stability AI emphasizes that this model series is entirely trained on fully licensed datasets. The company previously signed strategic cooperation agreements with Warner Music Group and Universal Music Group to jointly develop a new generation of responsible AI music creation tools. Recently, Ethan Kaplan, former Chief Digital Officer of Universal Audio, officially joined Stability AI to lead the professional music product business. These moves have established a certain copyright compliance barrier for Stability AI in the AI music generation field, contrasting with peers like Suno and Udio who are currently facing copyright lawsuits.
Looking at the product evolution trajectory, Stability AI has undergone multiple iterations in the audio domain. Stable Audio was launched in 2023, upgraded to version 2.0 in 2024 with the addition of audio-to-audio editing features, and Stable Audio 2.5 was released in 2025 for enterprise-level applications, supporting brand-customized sound effects and audio inpainting. The release of version 3.0 marks its audio models entering a new phase of commercial application in terms of generation length, musical structure control, and multi-specification deployment capabilities.
Brand Studio is positioned by Stability AI as a "brand-powered, end-to-end creative production platform," with its core philosophy being to allow brand teams to "lock" visual and sonic guidelines into the AI workflow, thereby maintaining brand consistency when producing content at scale. The platform integrates Stability AI's image generation, video generation, and audio generation capabilities, allowing users to complete the full chain of creative production from graphic design to video scoring without switching between multiple tools.
Founded in 2019 and headquartered in London, UK, Stability AI is one of the representative companies in the open-source generative AI field. Its Stable Diffusion text-to-image model has a broad developer ecosystem globally. The company's current CEO is former Weta Digital CEO Prem Akkaraju, its Executive Chairman is Napster co-founder Sean Parker, and its board members include renowned director James Cameron. As of September 2025, the company was valued at approximately $1 billion, with cumulative funding of around $400 million, primarily from investors including Coatue Management, Lightspeed Venture Partners, Greycroft, and Sound Ventures.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
