GoogleGoogle·🎡 Music Generation·New

Lyria 3 Pro

anonymized
Try on Venice.ai β†—
Quick reference
Lyria 3 Pro β€” TLDR
  • 🎡 Generates full, structured songs up to three minutes long.
  • 🎀 Supports vocals, custom lyrics, and multi-language generation across genres.
  • πŸ”§ Specify explicit intros, verses, choruses, and bridges.
  • πŸ†• Adds song-structure control beyond earlier Lyria releases.
  • 🎯 Builds complete compositions from a single text prompt.
  • 🌐 Available through Google Cloud Vertex AI.
  • πŸ”’ Outputs carry Google's SynthID audio watermarking.
  • 🏒 Google's full-length music generation tier, above the shorter clip variant.
πŸ’° Pricing
β€”
πŸ“… On Venice since
May 22, 2026
12 days ago
Provider

Google is an American multinational technology corporation and one of the world's most valuable brands. A subsidiary of parent company Alphabet Inc., Google operates across search, cloud computing, consumer electronics, and artificial intelligence. Its…

Read full profile β†’
25 models on Venice
10 text Β· 8 video Β· 2 image Β· 2 inpaint Β· 1 music Β· 1 embedding Β· 1 tts
Since Oct 15, 2024

About this model

Lyria 3 Pro is Google's full-length music generation tier, positioned to produce complete compositions rather than the short outputs of the Lyria 3 clip variant. From a single text prompt, it creates songs running up to roughly three minutes, with vocals, lyrics, and full instrumental arrangements across multiple genres and languages.

Its defining advance over earlier Lyria releases is structural awareness. Where prior clip-oriented versions offered limited control over song form, Lyria 3 Pro lets you explicitly specify sections such as intros, verses, choruses, and bridges, making it suited to longer pieces with more complex transitions and coherence across multiple parts.

The model sits within Google's broader generative media lineup alongside video models like Veo 3.1 Full Quality, image tools such as Nano Banana Pro, and text models including Gemini 3.1 Pro Preview.

Lyria 3 Pro is accessible through Google Cloud's Vertex AI platform, and its generated audio is embedded with Google's SynthID watermarking so that outputs can be identified as AI-generated. As with Google's other generative media tools, it is offered for developers and creators building music into their own products and workflows.

This About section is AI-generated from public sources (Claude Opus 4.8), with no human editing. It may contain inaccuracies β€” verify critical details against the sources listed above.

Data sources: Venice API Β· HuggingFace Β· Wikipedia β€” enrichment updated 5d ago