MiniMax Music 2.6
MiniMax Music 2.6 is a high-fidelity AI music model that generates full songs with realistic vocals and instrumentation from text prompts, featuring precise BPM/key control, auto lyrics, and instrumental-only mode.
Overview
What is MiniMax Music 2.6
MiniMax Music 2.6 is MiniMax's latest AI music generation model, released April 10, 2026. It creates full-length songs (up to ~6 minutes) with vocals and instrumentation from a text prompt and optional lyrics, supporting auto-generated lyrics, BPM/key accuracy, and instrumental-only output with high-fidelity audio up to 44.1kHz/256kbps.
Running it privately on Venice
On Venice, MiniMax Music 2.6 runs under an anonymized privacy tier—your prompts are not stored, profiled, or used for training. This means you can generate music with full creative freedom, knowing your inputs remain private and uncensored. The model is available on-demand with no subscription, so you only pay per generation.
Assessment
Strengths and limitations
- 99%+ accurate BPM and key control: specify 'E minor, 90 BPM' and the output matches.
- Realistic vocal timbre with natural breathing, vibrato, and pitch transitions, making vocals indistinguishable from studio recordings in demos.
- Full structural control using 14+ section tags like [Verse], [Chorus], [Bridge], [Drop], enabling precise song arrangement.
- Auto-generated lyrics from prompt when lyrics_optimizer is enabled — no manual writing needed.
- Instrumental-only mode with rich genre-specific mixing and high-fidelity output up to 44.1kHz/256kbps.
- Supports both custom lyrics input and fully AI-generated lyrics, giving flexibility in creative workflow.
- Proprietary and closed: no open weights, so self-hosting or fine-tuning is not possible.
- No support for image-guided music generation or negative prompts, unlike some rivals like Google Lyria 3.
- Higher per-generation cost compared to earlier versions and some competitors, at $0.18 per track.
- No multilingual vocal optimization benchmarked: performance may vary across non-Chinese/English languages.
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.”
Specifications
Datasheet
- Maker
- MiniMax
- Released
- April 10, 2026
- Modality
- Text-to-music, vocal & instrumental generation
- Max duration
- ~6 minutes
- Audio quality
- Up to 44.1kHz / 256kbps / WAV
- Structure control
- 14+ tags: [Intro], [Verse], [Chorus], [Hook], [Drop], [Bridge], etc.
- Lyrics handling
- Custom input or auto-generated via lyrics_optimizer
- Instrumental mode
- Yes — set is_instrumental to true
- Open weights
- No — proprietary
- Privacy on Venice
- Anonymized — prompts not stored
- Available on Venice since
- Apr 2026
- License
- Proprietary
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": "minimax-music-v26",
"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
Flat per-image pricing on Venice: $0.18 per generation.
New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.
Alternatives
How it compares
| Model | Best for | Max duration | Vocal realism | Open weights | Price (Venice) |
|---|---|---|---|---|---|
| MiniMax Music 2.6 | Latest version with BPM/key accuracy and auto-lyrics. | ~6 minutes | High — natural breathing, vibrato | No | $0.18 / track |
| MiniMax Music 2.5 | Same price but less precise BPM control and no auto-lyrics. | ~4 minutes | High | No | $0.18 / track |
| Lyria 3 Pro | Shorter tracks but lower cost and legal-safe training data. | 3 minutes | Configurable timbre | No | $0.10 / track |
| ElevenLabs Music | Higher cost and less music-specific structure control. | — | Strong vocal synthesis | No | from $0.69 / track |
Latest version with BPM/key accuracy and auto-lyrics.
Use cases
What it is good for
- 01Creating original songs with vocals and full instrumentation for indie artists and content creators.
- 02Generating background tracks for videos, games, or podcasts with precise mood and tempo control.
- 03Producing AI covers or style emulations for performance or educational use.
- 04Rapid prototyping of musical ideas from text descriptions without needing production skills.
- 05Building instrumental tracks for meditation, lo-fi, or ambient content with consistent structure.
Prompting
Getting better results
Include key and BPM in the prompt (e.g., 'E minor, 90 BPM') for accurate musical alignment.
Use structure tags like [Verse], [Chorus], [Bridge] to define song layout — they are respected by the model.
Enable lyrics_optimizer for fully AI-generated lyrics from a style prompt — ideal for ideation.
Set is_instrumental to true and omit lyrics to generate backing tracks or ambient pieces.
Specify vocal characteristics (e.g., 'male vocal, raspy, emotional') to guide timbre and delivery.
Version history
Introduced full song generation.
Improved realism and structure.
Added instrumental track support.
Current — BPM/key control, auto-lyrics, faster streaming.
FAQ
Frequently asked questions
MiniMax Music 2.6 is an AI music generation model released by MiniMax on April 10, 2026. It creates full-length songs up to ~6 minutes with vocals and instrumentation from a text prompt and optional lyrics, featuring realistic vocals, BPM/key control, and auto-generated lyrics.
On Venice, MiniMax Music 2.6 costs $0.18 per generation (one audio file), with no subscription or minimum spend required. You pay only for what you generate.
No. MiniMax Music 2.6 is a proprietary model — it is not free to use at scale and does not have open weights. It is not available for self-hosting or fine-tuning.
Yes. By setting the is_instrumental parameter to true, you can generate high-quality instrumental tracks without vocals. A style prompt is still required.
Yes. When the lyrics_optimizer parameter is enabled, MiniMax Music 2.6 can generate lyrics from just a style prompt, allowing you to create complete songs without writing any lyrics manually.
It supports up to 44.1kHz / 256kbps in WAV, MP3, or other formats, delivering studio-grade fidelity suitable for professional use in music, video, or broadcast.
Songs can be up to approximately 6 minutes long, though most generated tracks range between 2 and 4 minutes depending on complexity and structure.
MiniMax Music 2.6 excels in vocal realism, structural control, and longer tracks (up to 6 minutes), while Lyria 3 Pro is cheaper at $0.10 per track and offers legal-safe generation but caps at 3 minutes. Choose MiniMax for full creative control, Lyria for cost-effective, compliant output.
Yes. On Venice, your prompts are anonymized and not stored, profiled, or used for training. You retain full privacy and creative sovereignty when generating music.
Run MiniMax Music 2.6 privately
No prompt logging. No data used for training.