All rankings / Music
Music · vocals
Blind listener preferences published by Carnegie Mellon University’s Music Arena. Vocal and instrumental results are separate populations.
9 entries · Music Arena score · Higher is better
Source publication: 2026-07-31. Retrieved 2026-11-05.
Source: Carnegie Mellon University · Music Arena · MIT.
Ranks use confidence bounds. Several models can share a rank; the rows within a shared rank are displayed by score.
| Significance rank | Model / evaluated variant | Music Arena score | 95% confidence interval | Votes |
|---|---|---|---|---|
| 1 | Riffusion FUZZ v1.0Producer.ai | 1,132 | 1,093 – 1,171 | 426 |
| 1 | Lyria 3 ProGoogle DeepMind | 1,107 | 1,077 – 1,137 | 581 |
| 1 | Riffusion FUZZ v1.1Producer.ai | 1,103 | 1,065 – 1,141 | 421 |
| 1 | Sonauto v3 PreviewSonauto | 1,102 | 1,056 – 1,148 | 157 |
| 4 | Lyria 3 ClipGoogle DeepMind | 1,019 | 977 – 1,061 | 243 |
| 5 | ElevenLabs Music v1ElevenLabs | 998 | 974 – 1,022 | 872 |
| 7 | Sonauto v2.2Sonauto | 944 | 919 – 969 | 824 |
| 7 | ACE-Step 1.5 Turbo (1.7B)ACE Studio / StepFun | 939 | 910 – 968 | 668 |
| 9 | ACE-StepACE Studio | 656 | 587 – 725 | 312 |
What this ranking can tell you
- The date is the end of the source’s evaluation window, not the date this page was fetched.
- These results cover the source’s tested model versions. They do not establish the quality of later releases.
- Confidence bounds are reconstructed from the source’s rounded score and ± interval. Small differences should not be interpreted as decisive.
- No audio is redistributed. A benchmark result does not grant rights to a model or its generated music.
Check the source
The original evidence is retained byte for byte. The downloadable board includes the source URLs, revision and SHA-256 hashes.
Download this board and provenance →