ElevenLabs Launches Music v2.5, Adds API Access and Free Tier for AI-Generated Songs
ElevenLabs has released Music v2.5, an updated version of its ElevenMusic generator, now available through both the app and API. The company says blind testing with nearly 48,000 comparison pairs showed listeners preferred v2.5 over the prior version, particularly for R&B, Hip-Hop, and orchestral genres.
ElevenLabs has released Music v2.5, an update to its ElevenMusic generation model, making it available through both the ElevenLabs app and API with new free and paid tier options.
According to ElevenLabs, the new version produces "fuller and more natural" sounding tracks compared to its predecessor. The company backs this claim with a blind listening test involving 47,885 comparison pairs, in which listeners reportedly preferred v2.5 outputs most of the time — with the strongest preference gains in R&B, Soul, Hip-Hop, Rock, and orchestral music.
Pricing and Access
ElevenLabs is rolling out two access tiers for Music v2.5:
- Free tier: 5 lossless downloads per day, with attribution required for any commercial use
- Pro tier: 400 lossless downloads per month
Exact subscription pricing for the Pro tier was not disclosed in the announcement. Music v2.5 is also accessible via the ElevenLabs API, and the prior v2 model remains available for existing users.
Users retain rights to tracks they generate, and commercial use is permitted depending on industry and purpose. However, ElevenLabs blocks downloads of tracks that closely mimic other artists' existing songs, and the company states that imitating specific musicians is not allowed under its usage policy.
Training Data and Licensing
ElevenLabs says its Music models, including v2.5, were trained on "licensed stems and music," though the company did not detail the specific sources or scope of that licensing. This positions ElevenLabs differently from competitor Suno, which faces an ongoing lawsuit from major record labels alleging it trained models on copyrighted material without consent from rights holders. Suno released new models of its own this week, according to the report.
ElevenLabs separately signed a licensing agreement with Universal Music Group. The company clarified that this deal applies only to future, separate products and does not cover Music v2.5 or its existing music generation tools.
What This Means
ElevenLabs is positioning licensing transparency as a competitive differentiator in AI music generation, an area where legal exposure has become a defining risk factor following the Suno litigation. By emphasizing "licensed stems and music" as training data — even without full disclosure — the company is signaling to enterprise and commercial users that Music v2.5 carries lower legal risk than rivals facing active copyright suits.
The free tier with attribution requirements and a capped Pro tier download limit suggests ElevenLabs is testing a freemium funnel rather than aggressively pricing for enterprise scale yet. The absence of published API token pricing indicates this may still be an early-stage rollout, with usage-based pricing likely to follow as adoption data comes in. The real test will be whether the blind-test preference for v2.5 translates into measurable market share against Suno and other AI music tools already embedded in creator workflows.
Related Articles
Perplexity Says It Runs End-to-End Engineering Systems on OpenAI's GPT-6 Astra
Perplexity says it has shifted core engineering workflows, including code changes and production monitoring, onto OpenAI's GPT-6 Astra model. The claim comes from an OpenAI-published case study with no independent benchmark data released.
Perplexity Deploys OpenAI's Astra Model for Autonomous Code and Systems Management
Perplexity is using an OpenAI model referred to as Astra to handle software changes, communications, and production monitoring with less frequent human check-ins. OpenAI published the case study; specific model specs and benchmarks have not been disclosed.
Augment Code Claims 4.5x Developer Output Increase From Internal 'Software Factory' of AI Agents
Augment Code says its internal 'software factory'—a network of specialized agents built on its Cosmos platform—drove a 4.5x increase in size-adjusted developer output and cut median PR merge time from 11.2 to 3.1 hours over nine months. The company frames this as evidence that once AI writes nearly all new code, the bottleneck shifts to review, verification, and incident response.
AWS Shows How to Build Interactive MCP Apps on Amazon Bedrock AgentCore
AWS published a technical walkthrough for building MCP Apps—interactive HTML widgets rendered inside AI hosts like ChatGPT and Claude—using Amazon Bedrock AgentCore's runtime and Gateway components. The reference implementation, a unicorn rental app, demonstrates host-agnostic rich UI delivered through a single MCP server.
Comments
Loading...