Introduction
The Suno v6 extend endpoint continues a song you already have. You choose the second where the new material should begin; the result keeps the original audio up to that point and writes a continuation after it, with new lyrics or as an instrumental. It suits songs that end too early, demos that need a second verse and tracks that need a proper outro. Each request returns two complete versions of the extended song.
Key features & highlights
The join holds the groove
We extended three songs from our own platform at three kinds of point: after a final chorus, after a mid-song chorus and at the very end. In all six returned versions the tempo after the join matched the tempo before it, and the beat grid ran straight through the join. The loudness change at the join was in the same range as the change the original song itself makes at that point, so the new section reads as part of the song.
You get the whole song back
The result is not a loose clip. It contains the original up to extend_at_seconds and the new section after it, ready to publish. In our tests the continuation added between 45 and 91 seconds. Because everything after the join point is replaced, the same endpoint can also rewrite an ending you do not like.
Same controls as the generator
The new section follows lyrics and prompt, and accepts the familiar Suno controls: style_weight, weirdness, variety, vocal_gender, a negative style list and saved personas.
Request parameters
| Parameter | Required | Description |
|---|---|---|
audio_url | Yes | The song to continue, as MP3 or WAV of up to 8 minutes, recorded or generated by you. at least 1 characters |
extend_at_seconds | Yes | The join point in seconds. Original audio is kept up to it and replaced after it; place it on a phrase or section boundary for the smoothest join. |
prompt | No | Style of the new section in up to 1,000 characters. Repeating the song’s own style keeps the continuation consistent; changing it makes a deliberate turn. at least 1 characters · at most 1000 characters |
lyrics | No | Words for the new section, with section tags such as [Verse 2] or [Chorus] that guide the form and stay silent. at least 1 characters · at most 5000 characters |
instrumental | No | Turn on to continue without vocals; lyrics and |
title | No | Metadata title for the extended song, up to 100 characters. at least 1 characters · at most 100 characters |
negative_prompt | No | Up to 500 characters of styles or sounds the new section should avoid. at least 1 characters · at most 500 characters |
vocal_gender | No | Raises the chance that the new section is sung by a female or male voice; match the original singer. |
style_weight | No | How strictly the new section follows prompt, from 0 to 1. |
weirdness | No | Room for surprises in the new section, from 0 (expected) to 1 (unusual). |
audio_weight | No | Balance between the uploaded recording and the style in prompt, 0–1 in steps of 0.01. |
variety | No | Spread between the two returned versions: 0 off, 1 normal, up to 4. |
persona_id | No | ID of a saved Suno persona to carry its voice or style into the new section; send |
persona_model | No |
|
Pricing
$0.072 / request
One flat price per extension request; each returns two versions, and the length of the new section does not change the price.
Use cases
Grow a short song into a full track
A 70-second idea becomes a release-length song by adding a second verse and a final chorus. Point extend_at_seconds at the end of the last chorus, write the new lines with section tags and keep the original style in prompt so the singer and band stay the same.
Write a new bridge or outro
Place the join after a chorus in the middle of the song and describe a bridge, a key change or a long instrumental outro. Everything after the join is replaced, so you can test several directions for the second half without re-recording the first.
Prompt guide & pro tips
Choose the join point carefully
Put extend_at_seconds on a phrase or section boundary, such as the last beat of a chorus, rather than in the middle of a sung word. Listen to the original around that point first: a join placed just before a natural breath or drum fill sounds the most seamless.
Keep the style, change the words
Copy the song’s original style description into prompt and put only the new lines in lyrics, with tags like [Verse 2] or [Bridge]. Set vocal_gender to match the original singer. To change direction on purpose, describe the new feel in prompt.
Compare the two endings
The two versions usually differ in how long the continuation runs and how it ends. Check the join at normal volume, then the final seconds, and keep the one whose ending fits the release.
API quick start
# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
-H "Authorization: Bearer $SPICY_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
-d '{
"model": "suno/v6/audio-extend",
"input": {
"audio_url": "https://example.com/reference.mp3",
"extend_at_seconds": 0
}
}'
# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
-H "Authorization: Bearer $SPICY_API_KEY"
See the API reference for authentication, the complete schema, polling, and error handling.


