Minimax Lyrics Generation API Documentation
Playground
Try it on WaveSpeedAI!MiniMax Lyrics Generation creates complete song lyrics from natural-language prompts, and can edit or continue existing lyrics with structured output including a title, style tags, and lyrics ready for music generation. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Features
MiniMax Lyrics Generation creates complete song lyrics or edits and continues an existing draft from natural-language instructions. It returns a song title, comma-separated style tags, and lyrics with structure tags that can be used directly with MiniMax Music 3.0.
Why Choose This?
-
Complete song lyrics
Turn a theme, mood, genre, or story idea into structured lyrics for a full song. -
Editing and continuation
Revise, rewrite, or continue existing lyrics while preserving the creative direction you provide. -
Music-ready structure
The generated lyrics can include tags such as[Verse],[Chorus],[Bridge], and[Outro]for downstream song generation. -
Useful metadata
Each result includes a song title and style tags alongside the lyrics. -
Simple fixed pricing
Every generation or edit request has the same price, regardless of prompt length.
Parameters
| Parameter | Required | Description |
|---|---|---|
| mode | Yes | write_full_song creates complete lyrics; edit edits or continues existing lyrics. Default: write_full_song. |
| prompt | No | Song theme, style, mood, story, or editing instruction. Maximum length: 2000 characters. Leave empty in full-song mode for a random song. |
| lyrics | No | Existing lyrics to edit or continue. Used only in edit mode. Maximum length: 3500 characters. |
| title | No | Optional song title. When provided, the output keeps this title unchanged. |
How to Use
- Choose a mode — use
write_full_songfor new lyrics oreditfor an existing draft. - Describe the song — provide the theme, genre, mood, story, voice, or editing direction in
prompt. - Add existing lyrics when editing — paste the draft into
lyricswhen usingeditmode. - Optionally set a title — provide
titlewhen it must remain unchanged. - Submit — receive a title, style tags, and structured lyrics.
Example Prompt
A cinematic synth-pop love song set in a neon city at midnight, dreamy female vocals, intimate verses, a memorable soaring chorus, bittersweet but hopeful, with vivid rain and streetlight imagery.
Pricing
Just $0.02 per request.
Output
The model returns:
song_title— generated or preserved song titlestyle_tags— comma-separated musical and vocal style tagslyrics— generated lyrics with song structure tags
Supported structure tags include [Intro], [Verse], [Pre-Chorus], [Chorus], [Hook], [Drop], [Bridge], [Solo], [Build-up], [Instrumental], [Breakdown], [Break], [Interlude], and [Outro].
Best Use Cases
- Songwriting ideation — create a complete first draft from a concept.
- Verse and chorus development — strengthen or continue unfinished sections.
- Lyric rewriting — change mood, genre, imagery, or point of view.
- Demo preparation — generate structured lyrics ready for AI music creation.
- Creative content — write themed songs for videos, ads, games, and social media.
Pro Tips
- Include the genre, mood, narrator, setting, imagery, and desired emotional arc.
- Ask for a clear verse/chorus/bridge structure when arrangement matters.
- In edit mode, explain exactly what to preserve and what to change.
- Set
titlewhen your project already has a fixed song name. - Pass the returned
lyricsinto MiniMax Music 3.0 to generate a complete song.
Related Models
- MiniMax Music 3.0 — Generate a complete song from a music prompt and lyrics.
- Mureka AI Generate Lyrics — Generate lyrics from a text prompt.
- Mureka AI Extend Lyrics — Continue an existing lyric draft.
Authentication
For authentication details, please refer to the Authentication Guide.
API Endpoints
Submit Task & Query Result
set -euo pipefail
export WAVESPEED_API_KEY="your-api-key"
REQUEST_BODY=$(cat <<'JSON'
{
"mode": "write_full_song"
}
JSON
)
# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
-X POST "https://api.wavespeed.ai/api/v3/minimax/lyrics-generation" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
-H "Content-Type: application/json" \
-d "${REQUEST_BODY}")
TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
printf 'Submission response did not contain a prediction id
' >&2
exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"
# 2. Poll until the prediction finishes.
while true; do
RESPONSE=$(curl --silent --show-error --fail-with-body \
"${RESULT_URL}" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}")
RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')
case "${STATUS}" in
completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
failed|cancelled|timeout|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
*) sleep 2 ;;
esac
doneParameters
Task Submission Parameters
Request Parameters
| Parameter | Type | Required | Default | Range | Description |
|---|---|---|---|---|---|
| mode | string | Yes | write_full_song | write_full_song, edit | Generation mode. Use write_full_song to create complete lyrics or edit to modify or continue existing lyrics. |
| prompt | string | No | - | Song theme, style, or editing instruction. Leave empty in write_full_song mode to generate a random song. | |
| lyrics | string | No | - | - | Existing lyrics to edit or continue. Only used when mode is edit. |
| title | string | No | - | - | Optional song title. When provided, the generated output keeps this title unchanged. |
Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data.id | string | Unique identifier for the prediction, Task Id |
| data.model | string | Model ID used for the prediction |
| data.outputs | array | Output values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed) |
| data.urls | object | Object containing related API endpoints |
| data.status | string | Task status. completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses. |
| data.created_at | string | ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”) |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |
Result Request Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| id | string | Yes | - | Task ID |
Result Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data | object | The prediction data object containing all details |
| data.id | string | Unique identifier for the prediction |
| data.model | string | Model ID used for the prediction |
| data.outputs | array<string | object> | Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model. |
| data.urls | object | Object containing related API endpoints |
| data.status | string | Status: completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses |
| data.created_at | string | ISO timestamp of when the request was created |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |