GoogleMediumUltra

Lyria 3.5

Google's latest Lyria 3.5 music generation model via fal.ai. Generates full-length songs up to a few minutes with vocals and lyrics; control length and structure with duration hints or section timestamps in the prompt. Supports English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese. Returns an MP3 file.

230 credits

per song

Newest Lyria generation — full-length songs up to a few minutes

Length and structure control via duration hints or [0:00-0:30] section markers

Vocals and lyrics in 8 languages including Korean and Japanese

Optional image as mood inspiration

Returns generated lyrics alongside the MP3

Run it right now

Test this model instantly in the Console Playground — no code required

Use with AI Assistant

Copy usage instructions for Claude, ChatGPT, or other AI

llms.txtllms.txt ↗

Quick Start

curl -X POST "https://api.core.today/v1/predictions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: cdt_your_api_key" \
  -d '{
  "model": "google/lyria-3.5",
  "input": {
    "prompt": "밝고 경쾌한 K-pop 일렉트로팝 회사 징글, 128 BPM, 힘찬 여성 보컬과 남성 코러스, 신스 브래스와 펀치 있는 드럼. [0:00-0:10] Intro: 신스 브래스 훅. [0:10-0:35] Verse: '아침부터 AI, 커피보다 진해 / 코딩하다 보니 벌써 새벽 두 시네! / 안 되는 건 없어, 일단 만들어 봐 / 상상이 현실 되는 마법 같은 회사!'. [0:35-1:05] Chorus: '코어! 닷! 투데이! / 오늘의 상상이 내일의 기술로! / 코어! 닷! 투데이! / AI도 놀라는 우리의 아이디어!'. [1:05-1:25] Outro: '뚝딱뚝딱 만들고, 반짝반짝 빛내고! / 기술을 넘어 경험으로! / 코어닷투데이, Hey!'. 1분 25초 길이."
  }
}'

Parameters

Parameter	Type	Required	Default	Description
`prompt`	string	Yes	-	Text prompt describing the music. Include genre, mood, instrumentation, tempo, vocals, and structure. Use duration hints ('a 2-minute track') or section markers like '[0:00-0:30] Intro' to control length and structure.
`image_url`	string	No	-	Optional image URL used as inspiration — the music matches the image's mood and theme

How to Provide File Input

There are 3 ways to provide files for the image_url parameter:

Recommended

Image URL

Pass a publicly accessible URL directly. With the Storage API (POST /v1/files/upload-url) the file uploads straight to S3 (50MB per file) and you pass the returned file_url.

{
  "model": "google/lyria-3.5",
  "input": {
    "prompt": "your prompt here",
    "image_url": "https://example.com/image.jpg"
  }
}

Direct Upload (Multipart)

Attach files directly to POST /v1/predictions/upload. No separate upload step, but the bytes pass through the API server so the whole request is capped at 10MB.

curl -X POST "https://api.core.today/v1/predictions/upload" \
  -H "X-API-Key: cdt_your_api_key" \
  -F "model=google/lyria-3.5" \
  -F 'input={"prompt":"your prompt here"}' \
  -F "file:image_url=@your_file.png"

See the File Upload docs for more upload methods including Presigned URLs.

Common Parameters

Common parameters used when calling POST /v1/predictions.

Parameter	Type	Required	Default	Description
`model`	string	Yes	-	Model identifier
`input`	object	Yes	-	Object containing the model-specific parameters from the table above
`output_folder`	string	No	-	Folder path for output files (max 256 chars, '..' not allowed)
`webhook_url`	string	No	-	Webhook URL to call on completion
`is_public`	boolean	No	false	If true, output files are also available via permanent public URLs

Examples

Core.Today company song

The same Korean company-song lyrics given to every music model — compare how each one sings them

curl -X POST "https://api.core.today/v1/predictions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: cdt_your_api_key" \
  -d '{
  "model": "google/lyria-3.5",
  "input": {
    "prompt": "밝고 경쾌한 K-pop 일렉트로팝 회사 징글, 128 BPM, 힘찬 여성 보컬과 남성 코러스, 신스 브래스와 펀치 있는 드럼. [0:00-0:10] Intro: 신스 브래스 훅. [0:10-0:35] Verse: '아침부터 AI, 커피보다 진해 / 코딩하다 보니 벌써 새벽 두 시네! / 안 되는 건 없어, 일단 만들어 봐 / 상상이 현실 되는 마법 같은 회사!'. [0:35-1:05] Chorus: '코어! 닷! 투데이! / 오늘의 상상이 내일의 기술로! / 코어! 닷! 투데이! / AI도 놀라는 우리의 아이디어!'. [1:05-1:25] Outro: '뚝딱뚝딱 만들고, 반짝반짝 빛내고! / 기술을 넘어 경험으로! / 코어닷투데이, Hey!'. 1분 25초 길이."
  }
}'

Structured two-minute pop song

Duration and structure hints in the prompt shape the whole song

curl -X POST "https://api.core.today/v1/predictions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: cdt_your_api_key" \
  -d '{
  "model": "google/lyria-3.5",
  "input": {
    "prompt": "An uplifting pop song with driving electric guitar, energetic drums, and a catchy chorus singing 'Rise up and shine today'. 120 BPM. Two minutes long with a verse-chorus-verse-chorus-bridge-chorus structure."
  }
}'

Korean ballad with timed sections

Korean lyrics with section markers — generated through the gateway

curl -X POST "https://api.core.today/v1/predictions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: cdt_your_api_key" \
  -d '{
  "model": "google/lyria-3.5",
  "input": {
    "prompt": "잔잔한 피아노 발라드, 72 BPM, 여성 보컬, 어쿠스틱 악기만. [0:00-0:15] Intro: 피아노 솔로. [0:15-0:50] Verse: '창가에 남은 빛이 너를 닮아'. [0:50-1:30] Chorus: '다시 부르면 돌아올까'. 1분 30초 길이."
  }
}'

Tips & Best Practices

1State the target length ('a 2-minute track') — without it the model picks the length

2Section markers like '[0:30-1:00] Verse: ...' give the most control over structure and lyric placement

3Korean, Japanese and Hindi vocals are supported — write the lyrics in that language

4Negative prompts are not supported — describe what you want (e.g. 'acoustic instruments only') instead of what to avoid

5Served by fal.ai only (no Replicate route yet), so there is no cross-provider fallback

Use Cases

Korean-language songs for videos, ads and short-form content

Structured songs with timed intro / verse / chorus sections

Theme songs and full-length background music

Model Info

ProviderGoogle

Version3.5

CategoryAudio & TTS

Price230 credits

API Endpoint

POST /v1/predictions

Try in Console

Related Models

ACE-Step

ACE Studio • 16 credits

Bark

Suno • 190 credits

Chatterbox

Resemble AI • 59 credits

Chatterbox Multilingual

Resemble AI • 17 credits

Back to Docs