ElevenLabs Launches Music v2 With Inpainting and Dense Lyrical Delivery

ElevenLabsElevenLabs

· Updated

ElevenLabs released Music v2, a foundational model upgrade that introduces precise inpainting controls and the ability to handle complex genre transitions mid-track. The update significantly lowers costs for commercial and developer tiers while improving vocal fidelity for fast-paced delivery like rap.

ElevenLabs, an AI audio research company, launched Music v2 as the engine for its music generation ecosystem. The model introduces inpainting—a control mechanism for regenerating specific track segments—and supports dense lyrical delivery like fast rap. It also enables complex mid-track genre shifts, including transitioning from opera to heavy metal.
Pricing (Creative)
Up to 40% lower cost
Pricing (API)
Up to 50% lower cost
Audio quality
44.1kHz
Commercial use
Permitted (licensed data)
Availability
ElevenMusic, ElevenCreative, ElevenAPI (soon)

This release shifts ElevenLabs from a voice-synthesis specialist to a direct competitor in the high-fidelity music market. By integrating non-musical sound effects and improving arrangement logic, the model enables complex compositions that were previously not possible. It follows the ElevenMusic platform launch by providing the control required for commercial production.

Music v2 is available now through ElevenMusic and ElevenCreative platforms, with an ElevenAPI release coming soon. Pricing for the creative suite has been reduced by up to 40%, while API costs are now up to 50% lower. The model is trained on licensed data and cleared for commercial use.

View the full update on elevenlabs.io
ElevenLabs
ElevenLabs
@ElevenLabs
X

Today we are launching Music v2. Better vocals, instrumentation, and arrangement across every genre, improved multilingual support plus capabilities that weren't possible before. https://t.co/mJYCZ24lgO

138retweets957likes
View on X

Still wondering? A few quick answers below.

Music v2 is the second generation of the foundational AI music model from ElevenLabs. It features significant improvements to vocal quality, instrumentation, and track arrangement. The model is designed to handle complex musical tasks like fast rap delivery and mid-track genre transitions, such as shifting from opera to heavy metal within a single song.

Inpainting is a new control capability that allows users to select a specific section of a generated track and regenerate only that part. This enables creators to tweak a bridge or a verse without changing the rest of the song, providing granular editing control that was previously not possible in generative audio workflows.

Yes, the model is trained on licensed data and is cleared for commercial use. On self-serve plans, users can use the music for most online and offline commercial purposes. Enterprise plans offer broader rights, including use in film, television, and large studio games, making it suitable for professional advertising and branded content.

Alongside the launch of the v2 model, ElevenLabs has reduced pricing for its commercial and developer tiers. The cost for ElevenCreative has been cut by up to 40 percent, while the ElevenAPI is now up to 50 percent lower. These changes apply to self-serve customers looking to integrate high-quality AI music generation into their products.

Music v2 is available now through the ElevenMusic platform for creating and remixing tracks, as well as the ElevenCreative suite for branded content. While the model is currently accessible via the web application, the ElevenAPI is coming soon. Developers can currently inquire for early access to integrate the model directly into their own products.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

ElevenLabs

See all AI news & updates from ElevenLabs →

Share this update