Minimax Music 2.5 API Documentation

Minimax Music 2.5 API Documentation

Playground

MiniMax Music 2.5 is a full-dimensional breakthrough in AI music generation with high-fidelity audio, humanized vocals, and precise creative control. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

MiniMax Music 2.5 is an advanced AI music generation model that creates complete songs from text prompts and lyrics. Describe your desired musical style and provide lyrics — the model generates full-length music with vocals, instrumentals, and professional production quality.

Why Choose This?

• Complete song generation Creates full songs with vocals and instrumentals from text descriptions.

• Lyrics support Input your own lyrics with structure markers (Verse, Chorus, etc.) for precise song composition.

• Style flexibility Supports a wide range of genres from cinematic orchestral to pop, rock, electronic, and more.

• Professional audio quality Multiple bitrate and sample rate options for studio-quality output.

• Prompt Enhancer Built-in tool to automatically improve your music descriptions.

Parameters

Lyrics Format

Use structure markers to guide song composition:

• (Instrumental intro) — Instrumental section without vocals
• (Verse) — Verse section
• (Chorus) — Chorus/hook section
• (Bridge) — Bridge section
• (Outro) — Ending section

Example:

(Instrumental intro with drums building up) (Verse) The sun rises on the broken ground Silence screams without a sound (Chorus) I stand alone against the gale Through the storm, I will prevail

How to Use

1. Write your prompt — describe the musical style, genre, mood, and instruments.
2. Enter your lyrics — provide song lyrics with structure markers.
3. Set audio quality (optional) — choose bitrate and sample rate.
4. Run — submit and download your generated song.

Pricing

Best Use Cases

• Original Music Creation — Generate complete songs for creative projects.
• Demo Production — Quickly prototype song ideas before full production.
• Content Soundtracks — Create custom music for videos, podcasts, and games.
• Songwriting Assistance — Hear your lyrics set to music instantly.
• Commercial Jingles — Produce short musical pieces for advertising.

Pro Tips

• Use the Prompt Enhancer to refine your style descriptions.
• Be specific about genre, instruments, tempo, and mood (e.g., “cinematic orchestral, epic, Hans Zimmer style, motivational”).
• Include production details like “wide soundstage, 8k audio quality” for enhanced results.
• Use structure markers in lyrics to control song arrangement.
• Higher bitrate (256000) and sample rate (44100) deliver best audio quality.
• For instrumental sections, use markers like “(Instrumental intro)” or “(Guitar solo)”.

Notes

• Both prompt and lyrics are required fields.
• Generated songs include AI-synthesized vocals matching the lyrics.
• Song length depends on lyrics length and structure.
• For best results, provide detailed style descriptions and well-structured lyrics.

Related Models

• MiniMax Speech 2.8 HD — High-quality text-to-speech generation.
• MiniMax Speech 2.8 Turbo — Fast text-to-speech generation.

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'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "lyrics": "Waves rise softly under the morning light",
  "bitrate": 256000,
  "sample_rate": 44100
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/minimax/music-2.5" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters