Google has made Lyria 3.5, the latest model in its Lyria music-generation family, available across the Gemini ecosystem. The rollout brings higher-fidelity music generation, genre and style controls, new creative templates, and developer access through the Gemini API. For teams producing videos, presentations, campaigns, or digital products, the practical change is a broader set of Google-native ways to create custom backing tracks, vocal songs, and other audio assets.
According to Google's official Lyria 3.5 announcement, the model is available globally through the Gemini app on web and mobile. Google is also extending access through Flow Music, Google AI Studio, Google Vids, and the Gemini API. That makes this more than an app feature. It is a platform-wide update that serves creators who want a quick starting point as well as developers building music generation into software and workflows.
Lyria 3.5 is designed to produce richer arrangements and more expressive vocals than earlier Lyria capabilities. Users can describe a genre or select one, choose vocal or instrumental output, and set the desired track length. Gemini templates provide another route for getting started, offering structured creative prompts for people who do not want to begin with a blank text field.
What Lyria 3.5 adds to Gemini music generation
The central advance is the combination of more detailed musical output with controls that map to common production decisions. Rather than treating music generation as a single prompt-and-output task, Google is exposing choices around genre, vocal presence, and duration. Those controls can help users shape an asset around its intended format, whether that is a backing track for a video, a brand jingle, or a ringtone.
The Gemini API documentation describes Lyria 3.5 as generating stereo audio at 44.1 kHz from text prompts or image prompts. It can create songs with structural elements such as intros, verses, choruses, and bridges, while also providing lyric timing. The documentation further says the model can adapt lyrics across languages.
For creators, templates may be the most immediately accessible part of the release. They can reduce the work of translating an idea into a detailed musical prompt. For developers, API access opens the possibility of creating product experiences that generate music from a customer brief, a visual asset, or a text description. The documentation indicates that API outputs can include both audio data and lyric text.
Key confirmed capabilities include:
- Genre selection or description to guide the musical direction.
- A choice between vocal and instrumental styles.
- Flexible creation of short or longer tracks, including full-length output use cases.
- Gemini templates intended to help users begin creative work faster.
- Text-to-music and image-to-music prompting through the Gemini API.
- Structured songs that can include intros, verses, choruses, bridges, and timed lyrics.
Google has not provided pricing details for Lyria 3.5 in the supplied announcement and documentation. Businesses evaluating the API should therefore check the current Gemini API terms and product access details before planning a production use case.
Lyria 3.5 and Lyria 3 Clip serve different jobs
Google's API materials distinguish the full Lyria 3.5 model from Lyria 3 Clip, a model for 30-second previews. The documentation suggests using Clip to refine a prompt before generating a full-length result with Lyria 3.5. This creates a practical iteration path: test the creative direction with a shorter output, then use the full model once the prompt is ready.
| Capability | Lyria 3 Clip | Lyria 3.5 |
|---|---|---|
| Primary output use | 30-second previews | Full-length music outputs |
| Suggested workflow | Prompt refinement | Final generation after refinement |
| Song structure and lyrics timing | Not specified in the supplied research | Supports structural coherence and lyric timing |
| Availability context | Gemini API documentation | Gemini app, Gemini API, Flow Music, AI Studio, and Google Vids |
What the rollout means for creative and product workflows
Lyria 3.5's availability across several Google products matters because it meets users in different parts of the creative process. A marketer or video producer can work in the Gemini app or Google Vids, while a developer can access the capability through Google AI Studio and the Gemini API. Artists and creative users also have an additional route through Google Flow Music.
For businesses, the useful question is not simply whether an AI-generated track sounds good. It is whether a generated asset can be created, reviewed, and used reliably in a specific workflow. The new controls can make briefs more concrete: a team can specify an instrumental track instead of a vocal one, guide the genre, and select an appropriate duration before generating a draft.
Potential uses supported by Google's stated examples include:
- Backing tracks for video content.
- Brand jingles.
- Ringtones.
- Music concepts generated from written or visual prompts.
The API's image-prompt support is particularly relevant to teams that already work from campaign visuals, storyboards, product imagery, or other creative references. It does not remove the need for editorial judgment. A generated result still needs a human review for fit, consistency with the intended message, and suitability for the final channel. But the ability to move from an initial brief to a tailored audio draft within the Gemini ecosystem could shorten early-stage experimentation.
The cross-product rollout also gives developers a clearer separation between prototype and production-oriented tasks. A short Clip output can be used to explore a direction, while Lyria 3.5 can be used for a more complete result once the prompt has been improved. That is a more deliberate process than repeatedly generating longer tracks without first validating the creative brief.
For companies considering generative music, the immediate opportunity is to identify narrow, repeatable situations where custom audio would add value but conventional production is too slow or resource-intensive. The model's track controls and API availability are useful only when they are connected to a defined content process, such as creating variations for video concepts or turning approved creative inputs into audio drafts.
Scalevise helps businesses turn new AI capabilities into practical processes, from choosing viable use cases to connecting models with existing tools and review steps. If your team wants to assess where generative audio can reduce manual creative work without disrupting quality control, our AI implementation consultancy can map the right workflow, technical approach, and rollout priorities. Request a consultation to discuss an AI adoption plan.
Frequently Asked Questions
Is Google Lyria 3.5 available now?
Yes. Google says Lyria 3.5 is available globally in the Gemini app on web and mobile, with access also extending to the Gemini API, Flow Music, Google AI Studio, and Google Vids.
What can Lyria 3.5 generate?
Lyria 3.5 can generate high-quality stereo music from text or image prompts. Users can select or describe a genre, choose vocal or instrumental styles, set track length, and create songs with structures such as verses, choruses, and bridges.
What is the difference between Lyria 3.5 and Lyria 3 Clip?
Lyria 3 Clip produces 30-second previews. Google's documentation suggests using it to refine prompts before using Lyria 3.5 for full-length music output.
Does Lyria 3.5 support developers?
Yes. Lyria 3.5 is available through the Gemini API's Interactions API. API outputs can include audio data and lyric text, and the model supports text and image prompting.
Has Google announced Lyria 3.5 pricing?
The supplied Google announcement and API documentation do not provide Lyria 3.5 pricing details. Users should review current Gemini API access and terms before planning a production deployment.
Conclusion
Google's Lyria 3.5 rollout makes higher-fidelity, controllable music generation available across Gemini's consumer, creative, and developer surfaces. Its genre, vocal, duration, template, and API options give users clearer ways to turn creative briefs into audio drafts. The most useful applications will be those where teams can connect generation to a defined review and production workflow.
Top comments (0)