Lyria 3 Clip Preview

Image inputAudio output

Music generation model for short 30-second clips, loops, and previews from text or image prompts

Released: 6mo ago
Pricing: Not published

Audio pricing is shown only when an audio billing unit is published. Transcription and speech rates are converted to per hour and per 1M characters.

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

No benchmark results have been matched to this model yet.

Specifications

Developer: Google DeepMind
Model family: Lyria
Type: Audio model
Predecessor: —
Successor: —
Released: March 25, 2026
Model ID: google/lyria-3-clip-preview
Openness: Closed
License: —
Input modalities: text, image
Output modalities: text, audio
Modality: text+image->text+audio
Hugging Face: —
Source freshness: Metadata synced Oct 8, 2026, 10:17 PM UTC

Additional source evidence

Source model: google/lyria-3-clip-preview
Source family: lyria
Last updated: 2026-03-25
Model context: 131K
Model output limit: 66K
Open weights: No
File attachments: Yes
Reasoning: No
Tool calling: No
Structured output: No
Temperature control: Yes

Complete source record (JSON)

{
  "id": "google/lyria-3-clip-preview",
  "name": "Lyria 3 Clip Preview",
  "description": "Music generation model for short 30-second clips, loops, and previews from text or image prompts",
  "family": "lyria",
  "last_updated": "2026-03-25",
  "attachment": true,
  "reasoning": false,
  "tool_call": false,
  "structured_output": false,
  "temperature": true,
  "modalities": {
    "input": [
      "text",
      "image"
    ],
    "output": [
      "text",
      "audio"
    ]
  },
  "limit": {
    "context": 131072,
    "output": 65536
  },
  "release_date": "2026-03-25",
  "open_weights": false
}

Infrastructure

Provider availability

All providers

Provider	Endpoint	Pricing (published units)	Uptime 1d	Data handling
Google AI Studio	google-ai-studio	—	100.00%	ZDR: No Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15

Provider data:OpenRouter

Text provenance

Gemini app/web

Google SynthID text watermarking

Google DeepMind says text generated in the Gemini app and web experience is watermarked with SynthID by adjusting token probabilities.

Official explanation