ElevenLabs Music v2.5
ElevenLabs' flagship song-generation model — richer melodies, more natural instruments, and layered arrangements built for commercial use.
curl https://api.venice.ai/api/v1/audio/queue \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "elevenlabs-music-v2-5",
"prompt": "An uplifting cinematic orchestral build with soaring strings"
}'
# Use the returned queue_id with https://api.venice.ai/api/v1/audio/retrieve.
# Call /audio/complete after downloading if needed.Overview
What is ElevenLabs Music v2.5
ElevenLabs Music v2.5 is ElevenLabs' flagship text-to-music model, released in September 2026 as the default for prompted and reference generation in ElevenMusic. It produces fuller, more layered songs with natural-sounding instruments, accepts audio references to steer style, and gives users rights to the tracks they create.
Using it anonymously on Venice
On Venice, Music v2.5 runs under the anonymized privacy tier — your prompts are not stored, profiled, or tied to a personal generation history. There is no subscription to maintain: you pay per track, from $0.69 for a 60-second clip, and only when you generate. The same frontier music model ElevenLabs ships, used privately and permissionlessly.
Specifications
Datasheet
- Maker
- ElevenLabs
- Modality
- Text-to-music, audio-reference generation
- Open weights
- No — proprietary
- License
- Proprietary
- Released
- September 11, 2026
- Privacy on Venice
- Anonymized — prompts not stored
- Available on Venice since
- Sep 2026
Assessment
Strengths and limitations
- Clear quality step up over Music v2: richer melodies, more layered arrangements, and instruments that sound closer to a live take.
- Strongest reported gains on vocal-led and acoustic-heavy genres — R&B, soul, hip-hop, rock, metal, orchestral, and cinematic music.
- Audio references (roughly 30 seconds) guide sound, production style, instrumentation, tempo, and mood without copying the upload.
- Curated and custom Finetunes let you lock in a consistent stylistic identity across generations.
- You hold rights to what you make, with commercial use available — a real differentiator in a category full of restrictive terms.
- Runs privately on Venice: zero retention of your prompts, pay per track instead of a monthly plan.
- Closed and proprietary: no open weights, so it cannot be self-hosted or fine-tuned outside ElevenLabs' tooling.
- The headline quality win is a vendor-reported blind test against ElevenLabs' own Music v2; no independent head-to-head against Suno v6 or Lyria exists yet.
- Downloads are blocked on tracks that reference other artists' songs, and imitating existing musicians is not allowed.
- On Venice it costs meaningfully more per track than lighter rivals like ACE-Step 1.5 or MiniMax Music 2.0.
Use cases
What it is good for
- 01Scoring ads, branded content, and video with licensed, commercial-use music.
- 02Vocal-led demos in R&B, soul, hip-hop, and rock where instrument realism matters.
- 03Orchestral and cinematic beds for film, trailers, and games.
- 04Style-matched catalogs: steer every track with the same audio reference or custom Finetune.
- 05Pulling stems from a generated track for mixing, remixing, or post-production.
Prompting
Getting better results
Name the genre, era, and instrumentation explicitly — the model's biggest gains are on vocal-led and acoustic-heavy styles, so be specific about vocal tone and live instruments.
Write your lyrics directly into the prompt and structure them with section tags like [Verse], [Chorus], [Bridge] to control song form.
Upload a ~30-second audio reference when you want the sound, tempo, and production style matched — it guides style, it does not remix the upload.
Set the duration deliberately: a 60-second clip is the cheapest tier at $0.69, and price scales with length up to 360 seconds.
Request multiple variants in one run and pick the best take instead of re-rolling the same prompt.
Ask for stems on the tracks you intend to mix — 2, 4, or 6-stem separation is documented, so plan for it before you finalize.
Samples
Sample outputs
Generated on Venice with our standard prompt suite — the same prompts we run through every model of this type, so you can judge it like-for-like.
“An uplifting cinematic orchestral build with soaring strings, warm brass, and a hopeful resolution.”
“A mellow lo-fi hip-hop beat with a soft jazzy piano loop, vinyl crackle, and a relaxed late-night mood.”
Alternatives
How it compares
| Model | Best for | Max length | Open weights | Price (Venice) |
|---|---|---|---|---|
| ElevenLabs Music v2.5 | Vocal-led songs & commercial use | Up to 360s | No | from $0.69 / track |
| ElevenLabs Music v2 | Proven all-rounder | Up to 360s | No | from $0.69 / track |
| ACE-Step 1.5 | High-volume, budget generation | — | No | from $0.03 / track |
| MiniMax Music 2.0 | Cheap quick sketches | — | No | $0.04 / track |
| Lyria 3 Pro | Google's premium tier | — | No | $0.10 / track |
Pick ElevenLabs Music v2.5 when the track needs to sound like a real production — full songs with vocals, layered arrangements, and rights you can use commercially. For bulk drafts on a budget, ACE-Step 1.5 or MiniMax Music 2.0 get you there for pennies.
API
Call it from your code
Venice exposes this model through the REST API. Queue a generation with the model id.
curl https://api.venice.ai/api/v1/audio/queue \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "elevenlabs-music-v2-5",
"prompt": "An uplifting cinematic orchestral build with soaring strings"
}'
# Use the returned queue_id with https://api.venice.ai/api/v1/audio/retrieve.
# Call /audio/complete after downloading if needed.Pricing
What it costs on Venice
Pay per track on Venice — price scales with length, from $0.69.
New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.
FAQ
Frequently asked questions
ElevenLabs Music v2.5 is ElevenLabs' flagship text-to-music model, released on September 11, 2026, and now the default for prompted and reference generation in ElevenMusic. It generates full songs from natural-language prompts, with richer melodies, more layered arrangements, and more natural-sounding instruments than Music v2.
Venice bills per track, with the price scaling by length: $0.69 for a 60-second clip, $1.38 for 120 seconds, $2.08 for 180 seconds, $2.76 for 240 seconds, $3.45 for 300 seconds, and $4.15 for 360 seconds. There is no subscription required.
Neither. It is a closed, proprietary ElevenLabs model with no open weights, so it cannot be self-hosted or fine-tuned independently. On Venice you pay per generated track rather than a monthly fee, and new accounts get free welcome credits to try it.
There is no independent head-to-head benchmark yet. ElevenLabs' documented evidence covers v2.5 beating its own Music v2 in a 47,885-pair blind test, with the widest gaps on vocal-led and acoustic-heavy genres. Suno v6 ships as a three-variant family with its own editing workflow. If instrument realism and documented commercial rights are the priority, v2.5 has the stronger published case.
v2.5 is the newer default: in ElevenLabs' blind test across 47,885 same-prompt pairs, listeners preferred v2.5 the majority of the time, especially on R&B, soul, hip-hop, rock, metal, orchestral, and cinematic material. Music v2 (released May 26, 2026) stays available and costs the same on Venice, so v2.5 is the default choice unless you prefer v2's character.
On Venice you can generate clips from 60 seconds up to 360 seconds (six minutes), with the per-track price scaling linearly with duration. You can also let the model choose a dynamic length when you don't need a fixed duration.
Yes. ElevenLabs states that you have rights to what you make on every plan, with rights attaching the moment a track is created. Commercial use terms vary by tier on ElevenLabs' own platform, and downloads are blocked on tracks that reference other artists' songs — review the current terms for your intended use.
Yes. You can upload a short audio reference (up to roughly 30 seconds) to guide sound, production style, instrumentation, tempo, and mood — it influences style rather than copying or remixing the upload. Curated genre Finetunes and custom Finetunes built from your own original audio are also supported.
Venice runs it under the anonymized privacy tier: your prompts are not stored, profiled, or used for training, and generations are not tied to a personal history. You can use it in the Venice app or through the Venice API while keeping your creative work off Big-Tech data pipelines.
Use ElevenLabs Music v2.5 anonymously
Venice does not store your prompts. Chat history stays in your browser.