Skip to main content
ModelRunner
google avatar

lyria-3/clip API

google/lyria-3/clip

Generate a 30-second song clip — vocals and lyrics included by default — from a single text prompt, as an MP3.

0.04
OpenAPI

Model Input

Input

Describe the music you want in plain language — genre, mood, instrumentation, tempo and key (for example, 'uplifting indie-folk with bright fingerpicked acoustic guitar, warm claps and a hopeful melody'). Output is a fixed ~30-second clip. Vocals and lyrics are generated by default; add 'instrumental only, no vocals' to get an instrumental. Supply your own words with a 'Lyrics:' section and steer structure with tags like [Verse], [Chorus] and [Bridge]. Shape the arrangement over time with timestamped cues such as '[0:00-0:10] soft intro, [0:10-0:30] full band'. Lyrics can be produced in English, German, Spanish, French, Hindi, Japanese, Korean or Portuguese — name the language in the prompt. Negative prompting is not supported; steer with positive description instead.

You need to be logged in to run this model and view results.
Log in

Model Output

Output

Loading
Generated in 15.919 seconds
Logs (1 lines)

Model Example Requests

Examples

lyria-3/clip API

lyria-3/clip is a music AI model by google. On ModelRunner it runs through a REST API or via MCP from any AI assistant, at $0.04 per audio clip.

POST https://queue.modelrunner.run/google/lyria-3/clip

cURL

# Submit a request to the queue. Input fields go at the top level of the
# body. The optional reserved "metadata" object holds your own flat string
# tags — stored on the request, never sent to the model; filter later with
# GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>.
curl -X POST https://queue.modelrunner.run/google/lyria-3/clip \
  -H "Authorization: Key $MRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A warm lo-fi hip-hop beat with mellow Rhodes piano, soft vinyl crackle and a relaxed boom-bap groove — instrumental o…",
    "metadata": {
      "project": "my-project"
    }
  }'
# → { "request_id": "...", "status_url": "...", "response_url": "..." }

# Poll status_url until "COMPLETED", then fetch the result
curl "https://queue.modelrunner.run/google/lyria-3/clip/requests/$REQUEST_ID/status" \
  -H "Authorization: Key $MRUN_API_KEY"
curl "https://queue.modelrunner.run/google/lyria-3/clip/requests/$REQUEST_ID" \
  -H "Authorization: Key $MRUN_API_KEY"

JavaScript

import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("google/lyria-3/clip", {
  input: {
    "prompt": "A warm lo-fi hip-hop beat with mellow Rhodes piano, soft vinyl crackle and a relaxed boom-bap groove — instrumental o…"
  },
});
console.log(result);

Python

import os
import requests

headers = {"Authorization": f"Key {os.environ['MRUN_API_KEY']}"}

submitted = requests.post(
    "https://queue.modelrunner.run/google/lyria-3/clip",
    headers=headers,
    json={
      "prompt": "A warm lo-fi hip-hop beat with mellow Rhodes piano, soft vinyl crackle and a relaxed boom-bap groove — instrumental o…"
    },
).json()

# Poll submitted["status_url"] until "COMPLETED", then:
result = requests.get(submitted["response_url"], headers=headers).json()

Input parameters

NameTypeRequiredDescription
promptstringyesDescribe the music you want in plain language — genre, mood, instrumentation, tempo and key (for example, 'uplifting indie-folk with bright fingerpicked acoustic guitar, warm claps and a hopeful melody'). Output is a fixed ~30-second clip. Vocals and lyrics are generated by default; add 'instrumental only, no vocals' to get an instrumental. Supply your own words with a 'Lyrics:' section and steer structure with tags like [Verse], [Chorus] and [Bridge]. Shape the arrangement over time with timestamped cues such as '[0:00-0:10] soft intro, [0:10-0:30] full band'. Lyrics can be produced in English, German, Spanish, French, Hindi, Japanese, Korean or Portuguese — name the language in the prompt. Negative prompting is not supported; steer with positive description instead.

Machine-readable: OpenAPI schema · llms.txt

Use lyria-3/clip from Claude & Cursor (MCP)

Point Claude Code, Claude Desktop, Cursor, or any MCP client at the ModelRunner MCP server and lyria-3/clip becomes a tool your assistant can call directly — it authorizes via OAuth (no API key in config) and runs this model with the run_model tool using the endpoint google/lyria-3/clip.

MCP client config (Claude Desktop, Cursor)

{
  "mcpServers": {
    "modelrunner": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.modelrunner.run/mcp"]
    }
  }
}

Claude Code

claude mcp add --transport http modelrunner https://mcp.modelrunner.run/mcp

Then ask your assistant, for example: “Run google/lyria-3/clip on ModelRunner to generate music”. MCP setup guide.

Model Details

Model Details

Lyria 3 Clip is Google DeepMind's text-to-music model for short clips. Describe the track you want in a single `prompt` — genre, mood, instrumentation, tempo and key — and it returns a 30-second MP3 (192 kbps, 44.1 kHz). Unlike earlier Lyria models it generates vocals and lyrics by default, writing and singing words that fit your brief unless you ask for an instrumental. That makes it a fast way to turn a plain-language idea into a finished-sounding snippet with a hook.

## Best for - Short brand jingles, logo stingers and intro/outro music from a one-line brief - Background music beds for video, ads, games and presentations - Social-audio clips and notification or UI sounds - Auditioning a genre, mood or song idea before a full production - Sung hooks and vocal snippets in English or seven other languages

## Choose another model when - You need a full song or a track longer than 30 seconds — use a longer-form music model with a duration input - You want guaranteed instrumental music — either prompt "instrumental only, no vocals" or use an instrumental-only music model - You want spoken narration rather than music — use a text-to-speech model - You need sound effects or foley rather than music — use a sound-effects model

## Tips - Write the `prompt` like a music brief: name the genre, mood, key instruments and tempo ("upbeat indie-folk, bright fingerpicked acoustic guitar, hopeful, 110 bpm") - Vocals and lyrics are on by default; add "instrumental only, no vocals" to suppress them - Supply your own words with a "Lyrics:" section and shape structure with [Verse], [Chorus] and [Bridge] tags - Arrange the clip over time with timestamped cues such as "[0:00-0:10] soft intro, [0:10-0:30] full band" - Lyrics can be generated in English, German, Spanish, French, Hindi, Japanese, Korean or Portuguese — name the language in the prompt

## Limitations - Public preview and not intended for production use yet; behavior and availability may change - Output is a fixed ~30-second clip with one clip per prompt — there is no duration control - Negative prompting is not supported; steer with positive description instead

To run via the ModelRunner JavaScript client: ```js import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("google/lyria-3/clip", { input: { prompt: "Uplifting indie-folk with bright fingerpicked acoustic guitar, warm claps, and a hopeful sung chorus", }, }); ```