Audio

ACE-Step Music Generator

Describe a style, add lyrics, and download a song with vocals. Free text to music on devices you connect.

ACE-Step by ACE Studio and StepFun; Qwen3 Embedding by Alibaba Cloud

Model licenses

ACE-Step v1.5 turbo DiT, lyric encoder and derived conditioning · MIT AND Apache-2.0

Retain MIT copyright/permission/disclaimer and applicable full Apache-2.0 terms, preserve upstream notices and identify modifications. Apache patent termination and trademark limitations apply.

Distributed license

Notice

ACE-Step v1.5 audio VAE decoder · MIT

Retain MIT copyright/permission/disclaimer, preserve upstream notices and identify modifications.

Qwen3 Embedding 0.6B text encoder and tokenizer · Apache-2.0

Retain full Apache-2.0 terms, preserve upstream notices and identify modifications. Apache patent termination and trademark limitations apply.

FreeOnlineUnlimitedACE-Step by ACE Studio and StepFun; Qwen3 Embedding by Alibaba Cloud API access

API reference

Submit via REST

Use your personal API key (msk_…) to create jobs for this workload from your own tools. Jobs started this way run on the same devices you connected on Your Devices — at least one has to be online.

Hide API details
POSThttps://cp.meshlabs.cloud/v1/organizations/jobs

Authenticate with X-Organization-Api-Key: msk_your_personal_api_key. The header name matches org keys; personal keys use the same endpoint.

Request body

{
  "workload_type": "ace-step-15-turbo-music",
  "job_spec": {
    "prompt": "A photorealistic lighthouse at dusk",
    "vocalLanguage": "en",
    "durationSec": 60,
    "shift": 3
  }
}

job_spec fields

FieldRequiredDefaultDescription
promptYes—Genre, instrumentation, mood, vocal character.
lyricsNo—Optional. Supports [verse] / [chorus] / [bridge] structure tags. Leave empty for an instrumental.
vocalLanguageNoenLanguage code for the lyrics, e.g. en, zh, ja, es. Ignored for an instrumental.
durationSecNo60How long the song is. 60 seconds is what runs when the job omits the field.
shiftNo3Timestep schedule. 3 favours fidelity, 1 favours adherence to the prompt.
seedNo—Omit for a random song; set to reproduce one.
bpmNo—Omit to let the model choose.
keyScaleNo—For example C major or A minor. Omit to let the model choose.
timeSignatureNo—Beats per bar, e.g. 4 for 4/4 or 3 for 3/4.

cURL

curl -X POST https://cp.meshlabs.cloud/v1/organizations/jobs \
  -H "X-Organization-Api-Key: msk_your_personal_api_key" \
  -H "Content-Type: application/json" \
  -d '{
  "workload_type": "ace-step-15-turbo-music",
  "job_spec": {
    "prompt": "A photorealistic lighthouse at dusk",
    "vocalLanguage": "en",
    "durationSec": 60,
    "shift": 3
  }
}'
GEThttps://cp.meshlabs.cloud/v1/organizations/jobs/{job_id}

Poll with the same API key until status is terminal (COMPLETED, FAILED, CANCELED, or MODERATION_BLOCKED). When content moderation is enabled, jobs may start in MODERATION_PENDING before entering the compute queue. Completed jobs include presigned download_url values on each output.

Create response

{
  "id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
  "workload_type": "ace-step-15-turbo-music",
  "status": "PENDING",
  "job_spec_json": {
    "prompt": "A photorealistic lighthouse at dusk",
    "vocalLanguage": "en",
    "durationSec": 60,
    "shift": 3
  },
  "work_units": [],
  "outputs": [],
  "created_at": "2026-06-08T12:00:00Z",
  "updated_at": "2026-06-08T12:00:00Z"
}

Completed response

{
  "id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
  "workload_type": "ace-step-15-turbo-music",
  "status": "COMPLETED",
  "job_spec_json": {
    "prompt": "A photorealistic lighthouse at dusk",
    "vocalLanguage": "en",
    "durationSec": 60,
    "shift": 3
  },
  "work_units": [
    {
      "id": "…",
      "index": 0,
      "status": "COMPLETED"
    }
  ],
  "outputs": [
    {
      "id": "…",
      "output_type": "<output type>",
      "filename": "<output filename>",
      "size_bytes": 524288,
      "download_url": "<presigned output download URL>"
    }
  ],
  "created_at": "2026-06-08T12:00:00Z",
  "updated_at": "<completion timestamp>"
}

Describe a style, add optional lyrics, and download a song with sung vocals — up to about four minutes of stereo audio. Free text to music on devices you connect. The first run downloads the model; later songs start faster.