ElevenLabs Music v2.5 genre strengths and how to test them
ElevenLabs says v2.5 is strongest on vocal-led and acoustic-heavy genres such as R&B, soul, rock and orchestral. Test your genre with three briefs.

Music Business Worldwide reports that ElevenLabs calls v2.5 strongest on vocal-led and acoustic-heavy genres: R&B, soul, hip-hop, rock, metal, orchestral and cinematic. It also reports that v2.5 was preferred across 47,885 prompt pairs. That is ElevenLabs's own figure; the only reliable test for your genre is a short A/B of your own.
What was reported
| Claim | Detail |
|---|---|
| Launch date | September 11 |
| Strong genres | R&B, soul, hip-hop, rock, metal, orchestral, cinematic |
| Preference test | 47,885 prompt pairs |
| Licensing | Merlin, Kobalt and Believe among existing partners; the UMG deal is separate from v2.5 training |
A three-brief test
Write one brief per genre you ship, using the seven axes the Sume docs use: emotion, genre, tempo in BPM, key, instruments, arc and era. Give the same brief to every engine, then listen blind.
curl -X POST https://api.sume.com/v1/music-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: genre-test-orchestral" \
-d '{
"model": "sume/music-auto",
"prompt": "Cinematic orchestral, 72 BPM, A minor. Strings and French horn, a piano motif at 0:15, building to a full-orchestra peak. Era: modern film score. Instrumental."
}'Cost of the test
On Sume's Music Router a generation is $0.125, so three genre briefs cost 3 x $0.125 = $0.375. At ElevenLabs's $0.15 per minute list price for the Music API, a one-minute track is $0.15 each.
Reading results
- Judge vocal tracks and instrumentals separately.
- Do not trust a BPM claim in provider lyrics output; count it yourself.
- Keep the brief fixed and change only the engine.
Sources
Related posts
More in Models
- Gemini 2.5 Flash Image shut down Oct 2: what to call on Sume instead
Google's gemini-2.5-flash-image, the original Nano Banana, was set to shut down Oct 2, 2026. Which Sume image ids to test as a replacement, and what to check.
- Gemini 3.8 Flash TTS tops Hume's VoiceEQ board: run your own test
Hume's blog lists Gemini 3.8 Flash TTS atop its Real-World VoiceEQ board. Why a vendor-run board is only a lead, and how to run a blind A/B on your script.
- Omni Flash 1.1: GA in the Gemini API, Preview on Agent Platform
Gemini API release notes record gemini-omni-1.1-flash as GA on Aug 27, 2026; an Agent Platform page title still says Preview. What to check first.
- Omni's 1M-token context: how many ten-second edit turns fit?
Omni lists a 1,048,576-token context and 5,792 tokens per video second. If prior clips stay in context, 18 ten-second turns fit; Sume edits are stateless.
Written by Sume