Skip to main content

SunoAudio modelActive

Suno v6-wild

Popularityno data yet

n/ano search or usage data

Summary

Suno v6-wild belongs to the same music family but leans more heavily into variation and unexpected results. The web app accepts text, audio, images, and video and turns them into music.

Suno lists v6-wild for Pro and Premier users. No official public API endpoint is documented, leaving the documented delivery limited to the web app.

All benchmark results

Every published result with its source, test setup and comparison models.

Show all 9 published benchmark results

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

PER (Show measurement, test conditions, and source)
Source value
7.45
Metric
phoneme error rate
Unit
%
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Lower is better
Sample
192
Source type
vendor-reported
Evaluator
YuE2 model card
Status
active
Retrieved at
2026-10-04
Methodology
WildSongBench snapshot dated September 12, 2026: 192 selected outputs from 192 prompts. Suno v6 Wild selected the lower-PER candidate from two generations per prompt; each candidate had four ASR passes. The system comparison is not matched for compute.
Context
The YuE2 model team published this comparison; it is not an independent evaluation.
Source YuE2 model card
Musicality (Show measurement, test conditions, and source)
Source value
5.5644
Metric
score
Unit
No unit provided
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
vendor-reported
Evaluator
YuE2 model card
Status
active
Retrieved at
2026-10-04
Methodology
WildSongBench snapshot dated September 12, 2026: 192 selected outputs from 192 prompts. Suno v6 Wild selected the lower-PER candidate from two generations per prompt; each candidate had four ASR passes. The system comparison is not matched for compute.
Context
The YuE2 model team published this comparison; it is not an independent evaluation.
Source YuE2 model card
SongBench average (Show measurement, test conditions, and source)
Source value
6.4195
Metric
score
Unit
No unit provided
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
vendor-reported
Evaluator
YuE2 model card
Status
active
Retrieved at
2026-10-04
Methodology
WildSongBench snapshot dated September 12, 2026: 192 selected outputs from 192 prompts. Suno v6 Wild selected the lower-PER candidate from two generations per prompt; each candidate had four ASR passes. The system comparison is not matched for compute.
Context
The YuE2 model team published this comparison; it is not an independent evaluation.
Source YuE2 model card
WildSongBench SongEval average (Show measurement, test conditions, and source)
Source value
4.2199125
Metric
mean of five dimensions from a separate automatic quality evaluator; source maximum 5
Unit
rating
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
independent-evaluator
Evaluator
YuE project, WildSongBench results
Status
active
Retrieved at
2026-10-17
Methodology
Independent evaluation. WildSongBench comparison snapshot 2026-09-12. The evaluator reports Suno v6 Wild on 192 prompts with one selected output per prompt. Public baselines and Suno v6 settings use two candidates per prompt, four ASR passes per candidate, then select the lower-PER candidate. This is a selected-output comparison, not one unselected generation or a matched-compute test. The source does not publish per-metric seeds or confidence intervals.
Context
The result comes from the YuE project’s comparison, not a Suno-run publication. Candidate selection uses lower PER after four ASR passes per candidate; it is not an unselected single generation. The comparison is not matched-compute, and the source makes no statistical-significance claim.
Source YuE project, WildSongBench results
WildSongBench SongEval musicality (Show measurement, test conditions, and source)
Source value
4.1716401042
Metric
SongEval musicality dimension; source maximum 5
Unit
rating
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
independent-evaluator
Evaluator
YuE project, WildSongBench results
Status
active
Retrieved at
2026-10-17
Methodology
Independent evaluation. WildSongBench comparison snapshot 2026-09-12. The evaluator reports Suno v6 Wild on 192 prompts with one selected output per prompt. Public baselines and Suno v6 settings use two candidates per prompt, four ASR passes per candidate, then select the lower-PER candidate. This is a selected-output comparison, not one unselected generation or a matched-compute test. The source does not publish per-metric seeds or confidence intervals.
Context
The result comes from the YuE project’s comparison, not a Suno-run publication. Candidate selection uses lower PER after four ASR passes per candidate; it is not an unselected single generation. The comparison is not matched-compute, and the source makes no statistical-significance claim.
Source YuE project, WildSongBench results
WildSongBench AudioBox production quality (Show measurement, test conditions, and source)
Source value
8.1785263047
Metric
AudioBox production-quality score; source maximum 10
Unit
rating
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
independent-evaluator
Evaluator
YuE project, WildSongBench results
Status
active
Retrieved at
2026-10-17
Methodology
Independent evaluation. WildSongBench comparison snapshot 2026-09-12. The evaluator reports Suno v6 Wild on 192 prompts with one selected output per prompt. Public baselines and Suno v6 settings use two candidates per prompt, four ASR passes per candidate, then select the lower-PER candidate. This is a selected-output comparison, not one unselected generation or a matched-compute test. The source does not publish per-metric seeds or confidence intervals.
Context
The result comes from the YuE project’s comparison, not a Suno-run publication. Candidate selection uses lower PER after four ASR passes per candidate; it is not an unselected single generation. The comparison is not matched-compute, and the source makes no statistical-significance claim.
Source YuE project, WildSongBench results
WildSongBench MuLan text-audio similarity (Show measurement, test conditions, and source)
Source value
0.4998925799
Metric
MuLan style-audio similarity score; source maximum 1
Unit
rating
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
independent-evaluator
Evaluator
YuE project, WildSongBench results
Status
active
Retrieved at
2026-10-17
Methodology
Independent evaluation. WildSongBench comparison snapshot 2026-09-12. The evaluator reports Suno v6 Wild on 192 prompts with one selected output per prompt. Public baselines and Suno v6 settings use two candidates per prompt, four ASR passes per candidate, then select the lower-PER candidate. This is a selected-output comparison, not one unselected generation or a matched-compute test. The source does not publish per-metric seeds or confidence intervals.
Context
The result comes from the YuE project’s comparison, not a Suno-run publication. Candidate selection uses lower PER after four ASR passes per candidate; it is not an unselected single generation. The comparison is not matched-compute, and the source makes no statistical-significance claim.
Source YuE project, WildSongBench results
WildSongBench AllMusicCaps text-audio similarity (Show measurement, test conditions, and source)
Source value
0.4316435247
Metric
AllMusicCaps style-audio similarity score; source maximum 1
Unit
rating
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
independent-evaluator
Evaluator
YuE project, WildSongBench results
Status
active
Retrieved at
2026-10-17
Methodology
Independent evaluation. WildSongBench comparison snapshot 2026-09-12. The evaluator reports Suno v6 Wild on 192 prompts with one selected output per prompt. Public baselines and Suno v6 settings use two candidates per prompt, four ASR passes per candidate, then select the lower-PER candidate. This is a selected-output comparison, not one unselected generation or a matched-compute test. The source does not publish per-metric seeds or confidence intervals.
Context
The result comes from the YuE project’s comparison, not a Suno-run publication. Candidate selection uses lower PER after four ASR passes per candidate; it is not an unselected single generation. The comparison is not matched-compute, and the source makes no statistical-significance claim.
Source YuE project, WildSongBench results
WildSongBench Q3O prompt control (Show measurement, test conditions, and source)
Source value
4.5897676885
Metric
prompt-weighted overall Q3O control score on the source-reported 0–5 scale
Unit
rating
Benchmark version
WildSongBench comparison snapshot 2026-09-12
Category
audio
Direction
Higher is better
Sample
192
Source type
independent-evaluator
Evaluator
YuE project, WildSongBench results
Status
active
Retrieved at
2026-10-17
Methodology
Independent evaluation. WildSongBench comparison snapshot 2026-09-12. The evaluator reports Suno v6 Wild on 192 prompts with one selected output per prompt. Public baselines and Suno v6 settings use two candidates per prompt, four ASR passes per candidate, then select the lower-PER candidate. This is a selected-output comparison, not one unselected generation or a matched-compute test. The source does not publish per-metric seeds or confidence intervals.
Context
The result comes from the YuE project’s comparison, not a Suno-run publication. Candidate selection uses lower PER after four ASR passes per candidate; it is not an unselected single generation. The comparison is not matched-compute, and the source makes no statistical-significance claim.
Source YuE project, WildSongBench results

All specifications

Architecture

Model class
Experimental music generation
Source: Model class

Input and output

Input
Text, audio, images, and video
Source: Input
Output
Music audio
Source: Output

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking
Unknown
no source
Watermarking technique
Not known
no source

Licence and access

Access
Suno web app, Pro or Premier
Source: Access
Availability
No official public API access is documented
Source: Availability
Shutdown
None announced
no source

Further details

Style
More varied and unpredictable than Suno v6
Source: Style

Sources and data date

Benchmark sources at a glance

Which independent evaluators tested this model. This is coverage, not a quality ranking.

No active result from the reviewed external evaluators can currently be attributed to this model. Other source evidence remains available below.

Without vendor claims and historical or disputed results; those stay labelled in the single values.

TypeEvidence and data date
Research statusResearch date October 4, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not published

per-metric confidence intervals and run-level seeds

Note

The pinned results provide point estimates and n=192 selected outputs but no confidence intervals or per-metric seeds.

Additional sourceSuno v6 release notes (retrieved October 3, 2026) · Suno v6 release notes · Editorial description reviewed October 3, 2026
Additional sourceSuno v6 help article (retrieved October 3, 2026) · Suno v6 help article · Editorial description reviewed October 3, 2026
Additional sourceYuE2 model card (retrieved October 4, 2026)
Additional sourceYuE project, WildSongBench results (retrieved October 17, 2026)
Additional sourceModel metadata (retrieved October 17, 2026)12