🎉 New Year Sale: Get 20% OFF on all plans — Use code NEWYEAR2026.

Upgrade now
diffrhythm-short thumbnail

ModelsLab/Diffrhythm-Short

diffrhythm-short
Generate full-length songs with vocals and instrumentals in 10 seconds using latent diffusion—supports English and Chinese, 44.1kHz stereo, up to 4m45s tracks, requires only lyrics and style prompts.
API PlaygroundAPI Documentation

API Endpoint URL

Base URL for all API requests to this endpoint.

https://modelslab.com/api/v6/voice/song_generator

API Authentication

Authentication requires a valid API key included in the request. Generate and manage your API keys from your developer dashboard. Include the key in the key parameter for all API requests.

Integration Examples

Production-ready code samples for API integration

{
"init_audio": "https://assets.modelslab.com/generations/020e94f6-2928-438d-bd41-5f981cda7aa4.wav",
"prompt": "Write a soft romantic song about two people meeting under the rain for the first time, in an acoustic guitar style.",
"lyrics": "[00:10.00]In the gentle rain, where our eyes first met[00:14.00]Two souls wandering, lost in silhouettes[00:18.00]With a smile that warmed the coldest night[00:22.00]We found our hearts in the soft moonlight",
"lyrics_generation": true,
"model_id": "diffrhythm-short",
"webhook": null,
"track_id": null,
"key": "YOUR_API_KEY"
}

SDKs

Official SDKs

Production-ready SDKs and client libraries for all major programming languages

API Parameters

Technical specifications for API request parameters.

Field NameParameterTechnical Description
Audio refrenceinit_audio reference audio file to Copy the style of music
promptpromptEnter topic of the song you want to generate. Only useful when Generate Lyrics is set true
LyricslyricsLyrics must be provided in LRC format (including timestamps and lyric lines). This field is required if lyrics_generation is set to false.
Generate Lyricslyrics_generationSet generate lyrics true if you want to generate lyrics with the help of prompt, set it false when you want to Enter your own lyrics
Model Idmodel_idID of the music generation model to use Available options: diffrhythm-short, diffrhythm-long
webhookwebhookURL to receive POST notification upon completion
track_idtrack_idID for webhook identification