What is Suno Extend?
Suno Extend is the endpoint for making a song longer. You give it a source track and a timestamp, and it writes new music that carries on from that point in the same style. The source can be a clip from an earlier Suno music generation, referenced by its clip_id, or any audio file you host, passed as audio_url. The required continue_at field sets the second where the new material begins, so you can keep a strong intro and rewrite everything after the first chorus, or add a third verse and an outro to a short demo. You steer the continuation either with a short description (Simple mode) or with your own lyrics and section tags (Custom mode). The job runs async like every other Suno operation on Unifically and returns the extended track when it finishes.
Key features of Suno Extend
Four things decide how an extend request behaves.
Start from any second with continue_at
continue_at is required and takes decimal seconds, so you can cut in exactly on a downbeat (for example 99.6). The source plays up to that point, and the new section takes over from there.
Suno clips or your own audio
Pass clip_id to continue a clip from a previous Suno generation, or audio_url to continue a file you host. The same request shape covers both, so one code path handles generated and uploaded material.
Describe it or write it
Simple mode takes a one-line brief in gpt_description_prompt. Custom mode takes the actual lyrics in prompt, with tags like [Verse 3] or [Outro], so you control what gets sung in the new section.
Version choice, instrumental flag and custom style
Pick chirp-hawk, chirp-hawk-wild or chirp-goose with mv, set make_instrumental for a vocal-free continuation, and add a title for the new version. Setting mv to chirp-custom with a custom_model of 6 to 24 reference tracks extends in the style of your own music.
Best for
Songs that end too early
Generated clips often stop before the song feels finished. Extend from the last strong bar and add the bridge and outro it is missing.
Rewriting the second half
Keep an intro you like and set continue_at just before the part that went wrong, so the new section starts there.
Lengthening music for video
Stretch a cue until it covers the full scene instead of looping it or composing a new piece.
Finishing your own demos
Upload a rough sketch through audio_url and let Suno write the next section in the same style.
Longer instrumental beds
With make_instrumental on, extend background music for podcasts, streams or games without vocals appearing.
Use cases
A song-writing app can let users pick a point on the waveform, send that position as continue_at, and show the extended version next to the original. A video editor can measure the length of a timeline, find the gap between the music and the cut, and extend the soundtrack until it fits. A game studio can extend a theme at a transition point to build a longer combat or exploration variant from the same motif. Teams that already generate songs with the Suno music generation API can chain the two calls: generate a clip, keep its clip_id, then extend it in a second request with a new verse in Custom mode.
API examples
Parameter details live at docs.unifically.com/models/audio/suno/extend.
curl -X POST https://api.unifically.com/v1/tasks \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "suno-ai/extend",
"input": {
"mv": "chirp-hawk",
"custom": false,
"clip_id": "69ad968a-23c4-40e0-9cdc-80f9649c5209",
"continue_at": 99.6,
"gpt_description_prompt": "Continue the epic chorus"
}
}'
The request returns a task_id. Poll GET /v1/tasks/<task_id> until the status is completed, or pass a callback_url to have the extended track sent to your server when it is ready.
FAQs
People also ask
It continues an existing piece of audio from a point you pick. You send continue_at in seconds plus either a clip_id from an earlier Suno generation or an audio_url for a file you host, and the job returns new audio that picks up the song from that point.
Either works. Pass clip_id for a clip that came out of a previous Suno job, or pass audio_url for any audio file you host. Send one of the two, not both.
It is a number in seconds and accepts decimals, so values like 99.6 are valid. Set it to the bar or phrase where you want the new section to start.
With custom set to false you describe the continuation in gpt_description_prompt, for example "Continue the epic chorus". With custom set to true you write the actual lyrics or structure in prompt, using section tags like [Verse 3] or [Bridge].
Three Chirp versions. chirp-hawk (v6), chirp-hawk-wild (v6 Wild, for more experimental results) and chirp-goose (v6 Mini, faster and lighter). You can also set mv to chirp-custom and pass a custom_model object with a name and 6 to 24 reference audio URLs to extend in the style of your own tracks.
Yes. Set make_instrumental to true and the continuation comes back without vocals, which is useful when you are lengthening a backing track or a score cue.
