Hugging Face's logo

Datasets:
StemSplitio
/
stem-separation-benchmark-2026

Tasks:
Audio-to-Audio
Modalities:
Tabular
Text
Audio
Formats:
parquet
Languages:
English
Size:
< 1K
Tags:
music
audio
source-separation
stem-separation
vocal-removal
karaoke
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
stem-separation-benchmark-2026 / README.md
StemSplit's picture
StemSplit
feat(v1.1): add mdx_net_inst_hq3 + fix htdemucs_6s other-stem eval
1bbb1b9 verified
preview code
|
Raw
History Blame Contribute Delete
10.7 kB
metadata 
license: cc-by-4.0
language:
  - en
pretty_name: StemSplit Stem-Separation Benchmark 2026
size_categories:
  - 1K<n<10K
task_categories:
  - audio-to-audio
tags:
  - music
  - audio
  - source-separation
  - stem-separation
  - vocal-removal
  - karaoke
  - demucs
  - spleeter
  - mdx-net
  - bs-roformer
  - benchmark
configs:
  - config_name: metrics_only
    data_files:
      - split: results
        path: metrics_only/metrics.parquet

StemSplit Stem-Separation Benchmark 2026

A reproducible head-to-head comparison of every popular open-source music source-separation model against the StemSplit production API, evaluated on the standard MUSDB18-HQ test split using BSS Eval v4 and a small set of CC-BY tracks for qualitative listening.

Built and maintained by the StemSplit team. Source code: scripts/hf-benchmark on GitHub.

Leaderboard (median SDR per stem)

model_id bass drums other vocals
htdemucs_ft 10.38 10.11 6.34 9.19
mdx_extra_q 11.42 11.49 7.67 9.04
htdemucs_6s 9.11 9.54 5.74 8.66
htdemucs 9.78 10.01 6.42 8.53
mdx_net_inst_hq3 — — — 5.81

Higher is better. SDR is computed with the museval reference implementation (BSS Eval v4) on 1-second windows, exactly the protocol used by SiSEC and the SDX challenges, so these numbers are directly comparable to results reported in the literature.

Configurations

metrics_only (the leaderboard)

A Parquet table — one row per (model, track, stem) — for the full 50-track MUSDB18-HQ test split. No audio is shipped here; MUSDB18 is a research-only corpus and we are not allowed to redistribute it.

Schema:

column type meaning
model_id string matches configs/models.yaml
track_id string folder name in musdb18hq/test/
stem string one of vocals, drums, bass, other
sdr_median float median Signal-to-Distortion Ratio (dB)
sdr_mean float mean SDR (dB)
isr_median float median Image-to-Spatial Ratio (dB)
sir_median float median Signal-to-Interferences Ratio (dB)
sar_median float median Signal-to-Artifacts Ratio (dB)
n_frames int number of evaluation windows
sample_rate int typically 44100
duration_s float input track length in seconds
wall_time_s float inference wall time
rtf float real-time factor (wall_time / duration)
peak_rss_mb int peak resident memory of the runner process
peak_mps_mem_mb int peak Metal allocation (Apple Silicon only)
host_chip string e.g. Apple M4 Pro
host_unified_memory_gb string unified memory size on the test machine
commit_sha string repo SHA the run was produced from 
from datasets import load_dataset
ds = load_dataset("StemSplitio/stem-separation-benchmark-2026", "metrics_only")
df = ds["results"].to_pandas()
df.groupby(["model_id", "stem"])["sdr_median"].median().unstack()

audio_samples (planned for v1.2)

A future config will ship a small set of CC-BY-licensed clips with reference and predicted stems, so you can A/B them in the dataset viewer. Sourcing genuinely commercial-friendly 4-stem multitracks in 2026 is non-trivial; see the v1.2 sourcing notes for the plan.

Models

v1.1 (this release)

All five models below produced complete results on the full 50-track MUSDB18-HQ test split.

id family notes
htdemucs Hybrid Transformer Demucs Facebook AI's default 4-stem model
htdemucs_ft Hybrid Transformer Demucs Fine-tuned variant, best vocal/instrumental separation in the Demucs family
htdemucs_6s Hybrid Transformer Demucs 6-stem model (adds piano + guitar). Its other-row SDR is now computed against piano + guitar + other summed back together — see the Changelog for details.
mdx_extra_q Demucs MDX MDX challenge winner, quantised, 4-model ensemble — best bass and drums of the lineup
mdx_net_inst_hq3 MDX-Net Vocal isolator running through audio-separator with the CoreML ONNX provider on Apple Silicon. Only the vocals row is reported (it does not produce drums/bass/other).

Planned for v1.2

id family why deferred
bs_roformer Band-Split Roformer Current SOTA but slow on Apple MPS due to operator fallbacks (~11 hr for the full test set). Will add once we run on CUDA.
mel_band_roformer Mel-Band Roformer Same reason.
spleeter_4stems Spleeter Legacy baseline; TensorFlow install is brittle on Apple Silicon.

See configs/models.yaml on GitHub for exact versions and command lines.

Changelog

v1.1 (current)

  • Added mdx_net_inst_hq3 (vocals-only MDX-Net via audio-separator with the CoreML ONNX execution provider).
  • Fixed htdemucs_6s other-stem evaluation. The 6-stem model splits piano and guitar out of other; the v1 eval pass compared its residual other output directly to MUSDB's other and got a misleading 0.22 dB. v1.1 sums the model's piano + guitar + other predictions back together before SDR, so htdemucs_6s is now directly comparable to its 4-stem siblings.
  • New prediction_components column on every row records the WAV files that were summed to form each stem (e.g. vocals or other+piano+guitar), so the aggregation is fully auditable.

v1

  • Initial release: 4 Demucs-family models × 50 MUSDB18-HQ tracks = 800 rows.
  • Known issues (both fixed in v1.1): missing mdx_net_inst_hq3, htdemucs_6s other-stem evaluation artefact.

What StemSplit uses internally

The StemSplit hosted API runs HT-Demucs under the hood — the same models you can benchmark above. Pick a quality tier and look up its row in the leaderboard:

StemSplit tier Model row in this benchmark When to choose it
FAST htdemucs Speed-priority previews and bulk processing
BALANCED (default) htdemucs_ft Best vocal separation per second of compute
BEST (6-stem) htdemucs_6s When you need piano + guitar separately

In other words: this dataset is also a benchmark of StemSplit's own quality. We didn't add a separate stemsplit_api row because it would just duplicate those numbers.

Use the StemSplit API

If you'd rather not stand up Demucs, ffmpeg, torchcodec, and a GPU yourself, the StemSplit API ships the same models with a single HTTP call. Pricing, quotas, and the full feature set are documented in our developer portal:

Resource URL
🏠 Developer landing stemsplit.io/developers
📘 Getting-started docs (auth, upload, polling) stemsplit.io/developers/docs
📑 API reference (every endpoint, every field) stemsplit.io/developers/reference
🧩 Integration guides (Zapier, n8n, Make, Pipedream, Discord, Audacity, DJ workflows, ...) stemsplit.io/developers/guides

Minimal example — submit a job, poll, download stems:

# 1. Submit
JOB=$(curl -sS -X POST https://stemsplit.io/api/v1/jobs \
  -H "Authorization: Bearer $STEMSPLIT_API_KEY" \
  -F "file=@song.wav" \
  -F "stems=4")
JOB_ID=$(echo "$JOB" | jq -r .id)

# 2. Poll until completed
while [ "$(curl -sS https://stemsplit.io/api/v1/jobs/$JOB_ID \
  -H "Authorization: Bearer $STEMSPLIT_API_KEY" | jq -r .status)" != "completed" ]; do
  sleep 3
done

# 3. Download every stem
curl -sS https://stemsplit.io/api/v1/jobs/$JOB_ID \
  -H "Authorization: Bearer $STEMSPLIT_API_KEY" \
  | jq -r '.stems | to_entries[] | "\(.key) \(.value)"' \
  | while read stem url; do curl -sSL "$url" -o "$stem.wav"; done

→ Get an API key at stemsplit.io/developers.

Reproducing the results

Everything runs locally on a Mac with Apple Silicon — no CUDA required.

git clone https://github.com/yourusername/musicai
cd musicai/scripts/hf-benchmark
uv venv --python 3.11 && source .venv/bin/activate
uv pip install -e .

# 1. Get the data (~22 GB extracted from a 21 GB Zenodo zip)
python -m src.download_musdb

# 2. Run every enabled model on every track
python -m src.run_all --continue-on-error

# 3. Score with BSS Eval v4
python -m src.eval_metrics

# 4. Assemble the HF dataset (and optionally push)
python -m src.build_dataset
python -m src.push_to_hub --create

Reference wall times measured on an Apple M4 Pro (24 GB unified memory), PyTorch 2.11 with the MPS backend, for the v1.1 lineup. Demucs models use MPS; MDX-Net runs through ONNX Runtime's CoreML execution provider.

Stage Wall time
Download MUSDB18-HQ from Zenodo 32 min
Separate (5 models × 50 tracks) ~2 h 30 min
Eval (museval BSS Eval v4) ~2 h 20 min
Build dataset < 1 s
Total ~5 h 20 min

Why we built this

We run StemSplit — a hosted stem-separation service — and we needed an honest, public, reproducible way to compare ourselves to the state of the art. So we open-sourced it.

Want to ship a separation product without standing up GPU infrastructure? The same models are one HTTP call away — see the Use the StemSplit API section above, or jump straight to the developer docs and API reference.

Licensing

  • This dataset (the metrics, the dataset card, and the CC-BY audio samples) is released under CC-BY-4.0 — please cite us if you use it.
  • The MUSDB18-HQ audio referenced by metrics_only.track_id is not redistributed here. Download it from Zenodo under its own terms.
  • Each separation model retains its own license; see the table above.

Citation 

@misc{stemsplit_benchmark_2026,
  title  = {StemSplit Stem-Separation Benchmark 2026},
  author = {StemSplit},
  year   = {2026},
  url    = {https://huggingface.co/datasets/StemSplitio/stem-separation-benchmark-2026}
}