Custom social soundtracks
Create a short song or instrumental that matches a photo, video, joke, memory, or short-form post.
Independent tool overview
Lyria 3 at a glance
Google DeepMind's music-generation family for creating short clips or full songs from text and image prompts across Gemini and Google's developer platforms.
Visit the official Lyria 3 site ↗
- Developer
- Google DeepMind
- Models
- Lyria 3 Clip and Lyria 3 Pro
- Maximum length
- 30 seconds or about 3 minutes
- Inputs
- Text and images; Gemini also supports video inspiration
- Developer status
- Public preview
- Audio output
- 48 kHz stereo in the API
Overview
What Lyria 3 is
Lyria 3 is Google DeepMind's generative music family. The original Gemini release creates 30-second tracks from text, photos, or video, while the newer Lyria 3 Pro can generate structured songs up to roughly three minutes with elements such as intros, verses, choruses, and bridges.
Consumers can use Lyria inside the Gemini app. Developers can work with Lyria 3 Clip and Lyria 3 Pro through Google AI Studio and the Gemini API, while organizations can access the models through Vertex AI. Google also integrates the family into products including YouTube Dream Track and Google Vids.
The models can generate vocals and lyrics or instrumental audio, respond to instructions for genre, mood, instruments, tempo, and structure, and accept an image as creative input. Every generated track carries Google's SynthID watermark.
Use cases
Who Lyria 3 is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Full-song experimentation
Use Lyria 3 Pro when a concept needs a longer arrangement with explicit sections and transitions.
Music-enabled applications
Add text-to-music or image-to-music generation to creative products through the Gemini API or Vertex AI.
Video background music
Generate an original soundtrack matched to the mood and structure of a creator or business video.
Capabilities
Core Lyria 3 features
Clip and full-song models
Lyria 3 Clip targets 30-second generations, while Lyria 3 Pro produces songs up to approximately three minutes.
Lyrics and vocals
Can write lyrics from the prompt and generate vocal tracks, or produce instrumental music when requested.
Creative controls
Prompts can specify genre, era, mood, instruments, vocal character, tempo, and song structure.
Multimodal prompting
Text and images can directly guide the music models; the Gemini experience can also derive inspiration from uploaded videos.
Multiple Google surfaces
Available across the Gemini app, Gemini API, Google AI Studio, Vertex AI, Google Vids, and selected creator experiences.
SynthID watermarking
Google embeds an imperceptible digital watermark in generated audio to help identify AI-created tracks.
Process
How the Lyria 3 workflow works
Step 1
Choose the surface
Use Gemini for an interactive consumer workflow, AI Studio for prototyping, or the API and Vertex AI for product integration.
Step 2
Describe the track
Specify the subject, genre, mood, instruments, vocal or instrumental preference, tempo, and any required sections.
Step 3
Add visual inspiration
Optionally upload an image—or a video in Gemini—to guide the atmosphere and lyrical direction.
Step 4
Generate and export
Create a short clip or Pro song, then download the result and verify that its creative and usage requirements fit the project.
Cost
Lyria 3 pricing and free plan
Gemini users can create music with plan-based limits, and Google AI Plus, Pro, and Ultra subscribers receive progressively higher access. Developer API generations are billed per successful request.
Gemini
Included with Gemini access
Music generation is available to eligible users in the Gemini app with usage limits.
- Users must be 18 or older
- Requires a supported Google account and Keep Activity enabled
- Paid Google AI plans provide higher limits
Lyria 3 Clip API
$0.04/song
Paid Gemini API preview pricing for a 30-second song generation.
- Model ID: lyria-3-clip-preview
- No API free tier listed
Lyria 3 Pro API
$0.08/song
Paid Gemini API preview pricing for a full-song generation.
- Model ID: lyria-3-pro-preview
- No API free tier listed
Vertex AI
Usage-based
Enterprise access in public preview through Google Cloud.
- Cloud billing and regional availability apply
- Confirm the current Vertex AI price before production budgeting
Pricing checked . Check current pricing at the source ↗
Assessment
Lyria 3 strengths and limitations
Where it stands out
- Offers both quick 30-second clips and longer, structurally aware songs
- Accepts visual references as well as detailed text direction
- Available in consumer, prototyping, API, and enterprise environments
- Low per-song Gemini API pricing for experimentation and scaled workflows
- SynthID provides a built-in signal that audio was generated with Google AI
What to consider
- The API models remain preview releases, so behavior, limits, and pricing can change
- Gemini generation limits vary by account and subscription and are shown in-product
- Prompts naming artists are treated only as broad inspiration rather than direct imitation
- Filters and similarity checks do not eliminate every copyright or rights-management risk
- Generated music still needs human review for quality, brand fit, licensing requirements, and platform rules
Compare
Lyria 3 alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Content Creator
Suno AI
A dedicated AI-song platform with a mature consumer creation and sharing workflow.
Explore Suno AI →Content Creator
Udio
A music-first alternative for generating, extending, and iterating on songs.
Explore Udio →Content Creator
Stable Audio 3.0
A better fit for teams that want an open-weight, licensed audio model family.
Explore Stable Audio 3.0 →Content Creator
Adobe Firefly Audio
Consider Adobe when commercially oriented audio generation inside a broader creative suite matters most.
Explore Adobe Firefly Audio →Questions
Lyria 3 FAQs
What is Lyria 3?
Lyria 3 is Google DeepMind's family of music-generation models. It creates original vocal or instrumental tracks from text and image prompts, with consumer access in Gemini and developer access through Google's AI platforms.
What is the difference between Lyria 3 Clip and Lyria 3 Pro?
Clip creates 30-second tracks. Pro creates full songs up to approximately three minutes and better follows requests for sections such as intros, verses, choruses, and bridges.
Can Lyria 3 generate lyrics?
Yes. It can generate lyrics from a prompt, follow supplied lyrical direction, or create instrumental audio when requested.
How much does the Lyria 3 API cost?
Google lists Lyria 3 Clip Preview at $0.04 per generated song and Lyria 3 Pro Preview at $0.08 per generated song. The API pricing page does not list a free tier for these models.
Can I use an image or video as a prompt?
The developer models accept text and image inputs. In the Gemini app, users can also upload photos or videos as inspiration for a track.
Does Google watermark Lyria music?
Yes. Google says all Lyria 3 and Lyria 3 Pro outputs include its imperceptible SynthID watermark.
Bottom line
Our Lyria 3 verdict
Lyria 3 has grown from a playful 30-second Gemini feature into a credible music-generation family spanning short clips, full songs, APIs, and enterprise deployment. It is especially compelling for teams already building in Google's ecosystem, but preview status and rights review still matter for production use.
Visit Lyria 3 website ↗Similar Consumer Tools
View all Consumer AI tools →
Claude Sonnet 4.6
Claude Sonnet 4.6 - Anthropic's upgraded mid-tier model rivaling Opus at 1/5 the cost with 1M-token context
Gemini 3.1 Pro
Gemini 3.1 Pro - Google's upgraded flagship model with SOTA reasoning gains
Qwen3-TTS CustomVoice 1.7B
Qwen3-TTS CustomVoice 1.7B - Alibaba's multilingual text-to-speech model with voice cloning using just three seconds of reference audio
Qwen3.5 Small
Qwen3.5 Small - Alibaba's tiny open-source models that rival AI systems 13x their size
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.