# lyria-3/clip

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

## Overview

- **Endpoint**: `https://queue.modelrunner.run/google/lyria-3/clip`
- **Model ID**: `google/lyria-3/clip`
- **Category**: music
- **Kind**: inference
- **Tags**: music, text-to-audio, text-to-music, google, deepmind, lyria, lyria-3, vocals, song

## Pricing

- **Price**: $0.04 per output

## Request Lifecycle

This model runs on the ModelRunner **asynchronous queue API** — a single POST does not return the output. Every call requires an `Authorization: Key $MODEL_RUNNER_KEY` header. Run three steps:

1. **Submit** — `POST https://queue.modelrunner.run/google/lyria-3/clip` with a JSON body holding the input fields at the top level. The body may also include a reserved top-level `metadata` object — a flat string map (max 16 keys, key ≤64 / value ≤512 chars) stored on the request for your own tagging. It is never sent to the model; filter your request history with `GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>` (exact key=value matches, AND-ed). The response carries request handles only (no output yet):

```json
{
  "status": "IN_QUEUE",
  "request_id": "<21-char id>",
  "status_url": "https://queue.modelrunner.run/google/lyria-3/clip/requests/<request_id>/status",
  "response_url": "https://queue.modelrunner.run/google/lyria-3/clip/requests/<request_id>",
  "cancel_url": "https://queue.modelrunner.run/google/lyria-3/clip/requests/<request_id>/cancel"
}
```

2. **Poll status** — `GET <status_url>` until `status` is `COMPLETED`. Possible values are `IN_QUEUE`, `IN_PROGRESS`, `COMPLETED`, `FAILED`, `CANCELLED`. A `FAILED` request responds with HTTP 400 and an `error` field.

3. **Read result** — `GET <response_url>`. Returns the finished request, including the generated `output`:

```json
{ "id": "<request_id>", "status": "COMPLETED", "output": ..., "input": ... }
```

The JavaScript and Python SDKs below perform steps 2–3 for you. In any language without an SDK (Swift, Go, Kotlin, etc.) you must implement the polling loop and the final result fetch yourself — see the cURL example for the full flow.

### Input Schema

- **`prompt`** (`string`, _required_):
  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.

### Output Schema

_No `Output` schema properties are available._

## Default Example

**Input**

```json
{
  "prompt": "A cinematic orchestral cue with soaring strings, deep timpani and a triumphant brass finale — instrumental"
}
```

**Output**

```json
"https://media.modelrunner.ai/2GfXdyCenedTHLsCFnPwx.mp3"
```
## Usage Examples

### cURL

The queue API is asynchronous: submit the request, poll `status_url` until it is `COMPLETED`, then read the result from `response_url`. Requires `jq`.

```bash
# 1. Submit the request (returns request handles, not the output)
SUBMIT=$(curl --silent --request POST \
  --url https://queue.modelrunner.run/google/lyria-3/clip \
  --header "Authorization: Key $MODEL_RUNNER_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "prompt": "A cinematic orchestral cue with soaring strings, deep timpani and a triumphant brass finale — instrumental"
}')

STATUS_URL=$(echo "$SUBMIT" | jq -r '.status_url')
RESPONSE_URL=$(echo "$SUBMIT" | jq -r '.response_url')

# 2. Poll until the request leaves the queue / in-progress state
while true; do
  STATUS=$(curl --silent --url "$STATUS_URL" \
    --header "Authorization: Key $MODEL_RUNNER_KEY" | jq -r '.status')
  echo "Status: $STATUS"
  case "$STATUS" in
    COMPLETED) break ;;
    FAILED|CANCELLED) echo "Request $STATUS"; exit 1 ;;
  esac
  sleep 1
done

# 3. Read the finished request, including the generated output
curl --silent --url "$RESPONSE_URL" \
  --header "Authorization: Key $MODEL_RUNNER_KEY"
```

### JavaScript

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

const result = await modelrunner.subscribe("google/lyria-3/clip", {
  input: {
  "prompt": "A cinematic orchestral cue with soaring strings, deep timpani and a triumphant brass finale — instrumental"
}
});

console.log(result.data);
```

### Python

```python
import asyncio
import modelrunner_ai

async def main():
    response = await modelrunner_ai.submit_async(
        "google/lyria-3/clip",
        arguments={
  "prompt": "A cinematic orchestral cue with soaring strings, deep timpani and a triumphant brass finale — instrumental"
}
    )
    result = await response.get()
    print(result["output"])

asyncio.run(main())
```

## Additional Resources

- [Playground](https://modelrunner.ai/models/google/lyria-3/clip)
- [OpenAPI Schema](https://modelrunner.ai/models/google/lyria-3/clip/openapi.json)
- [LLM Instructions](https://modelrunner.ai/models/google/lyria-3/clip/llms.txt)