# Demucs Stem Separation

Edit and transform audio with Demucs Stem Separation from fal through one Spiral API key. Splits a song into separate stems such as vocals, drums, bass and other, with a six-stem model that also splits guitar and piano. It runs in the background and returns audio. Pay as you go, with no subscription.

- Provider: fal
- Category: Audio
- Runs: In the background
- Returns: audio
- Released: 2025-10-27
- Tool id: `fal` / `fal-ai/demucs`
- Page: https://spiral.ai/tools/fal/fal-ai/demucs

## Pricing

$0.00091 per second.

## Limits

- Longest queue wait: 1 hour
- Run deadline: 30 minutes
- Longest run, end to end: 255 minutes
- Largest inline input: 1 MB
- Largest input file: 2 GB
- Largest output: 5 GB

## Input

| Field | Kind | Allowed | Default | Description |
| --- | --- | --- | --- | --- |
| `audio_url` (required) | URL or file |  | — | URL of the audio file to separate into stems |
| `model` | one of | htdemucs, htdemucs_ft, htdemucs_6s, hdemucs_mmi, mdx, mdx_extra, mdx_q, mdx_extra_q | htdemucs_6s | Demucs model to use for separation |
| `stems` | list |  | ["vocals","drums","bass","other","guitar","piano"] | Specific stems to extract. If None, extracts all available stems. Available stems depend on model: vocals, drums, bass, other, guitar, piano (for 6s model) |
| `segment_length` | number (whole) | 1 to … | — | Length in seconds of each segment for processing (minimum 1). Smaller values use less memory but may reduce quality. Default is model-specific. |
| `shifts` | number (whole) | 1 to 10 | 1 | Number of random shifts for equivariant stabilization. Higher values improve quality but increase processing time. |
| `overlap` | number | 0 to 1 | 0.25 | Overlap between segments (0.0 to 1.0). Higher values may improve quality but increase processing time. |
| `output_format` | one of | wav, mp3 | mp3 | Output audio format for the separated stems |

## Example

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/demucs",
    "input": {
      "audio_url": "https://storage.googleapis.com/falserverless/model_tests/audio-understanding/Title_%20Running%20on%20Fal.mp3"
    }
  }'
```

Try it in the Playground: https://spiral.ai/start?provider=fal&model=fal-ai%2Fdemucs

## Common questions

### Demucs Stem Separation

**How much does Demucs Stem Separation cost?**

$0.00091 per second. You pay as you go from one Spiral balance, with no subscription, and a failed run is not charged.

**How do I use Demucs Stem Separation through an API?**

Send `POST /v1/runs` with `"provider": "fal"`, `"model": "fal-ai/demucs"` and the input fields on this page, using one Spiral API key. It runs in the background: poll the run, or name a webhook to be told when it finishes.

**Do I need an account with fal to use Demucs Stem Separation?**

No. Spiral runs it on its own fal account and charges your Spiral balance.

**Can my AI agent use Demucs Stem Separation?**

Yes. Agents find and run it through Spiral's MCP server, CLI or agent skill, with the same API key.

