What is Suno v6
Suno v6 is AI music generation platform Suno The new-generation model series launched includes the flagship model v6 (for Pro/Premier users), the experimental model v6-wild, and the free v6-mini. Compared with the previous generation, the model's generation speed is greatly improved, and it supports natural language editing of specific parts of a song, multi-audio-source mixing, sample extraction, and generating songs from audio and images. The company also announced partnerships with Warner Music and BMG, providing layered model suites for different creator groups.
Main features of Suno v6
- Extremely fast generation: After entering a prompt, a complete song can be output within seconds, with generation speed far ahead of the previous generation.
- Natural language local editing: You can regenerate only a specific section of a song through a prompt or by specifying a certain word in the lyrics.
- Multimodal reference generation: Supports using text, images, or even videos as reference materials to create new songs.
- Sample extraction: Can separate a single instrument from a song and build a new beat around it.
- Licensed Remix (planned): A remix program for a catalog that artists voluntarily join will be launched, allowing fans to Remix their works while artists receive a share of the revenue.
- Model tiers: v6 (flagship) and v6-wild (experimental) are available to Pro/Premier paying users, while v6-mini is free and open to everyone.
Technical principles of Suno v6
- Training from scratch based on licensed data: The model is a completely new model trained entirely from scratch. The training data comes only from the catalogs of licensed partners such as Warner Music, BMG, and Believe, and does not include data from Universal and Sony; partner record companies can choose whether to participate in training and receive corresponding compensation.
- Preference feedback optimization mechanism: Suno adopts an approach similar to RLHF: by default, two versions are generated for each song, and the behavioral signal of which one the user keeps is used to continuously optimize the model's preferences, gradually aligning the output with popular listening tastes.
- Multimodal conditional generation architecture: The model uniformly encodes heterogeneous inputs such as text, images, and videos into conditional signals to guide the audio generation process, enabling cross-modal creation such as "one image generates one song."
- Local regeneration (Inpainting) technology: Drawing on ideas from image restoration, v6 can mask and locate a time segment of a song and regenerate audio only within the specified section while keeping the rest unchanged, supporting natural language editing and word-level lyric modification features.
- Source separation and sample extraction: Built-in stems separation capability can break a mix down into independent tracks such as vocals, drums, and bass, and supports precisely extracting a single instrument sample from a track for secondary creation. Separation quality is cleaner and artifact-free compared with the previous generation.
How to use Suno v6
- Register an account: Visit Suno official website to register and log in. Free users can use v6-mini, while subscribing to Pro/Premier unlocks the flagship model v6 and the experimental v6-wild.
- Enter a prompt to generate: Describe the song style, theme, or paste lyrics in the creation box, then click generate to get the finished product within seconds.
- Choose a version: By default, two versions are generated for each song. Listen and keep the version you prefer; your choice will help optimize the model.
- Local editing: For passages you are not satisfied with, use natural language instructions or select a word in the lyrics to regenerate only that part.
- Multimodal creation: Upload images or videos as references and let the model generate soundtracks or songs based on them.
- Sample extraction: Use the track separation feature to extract a single instrument and build a new beat or remix around it.
Core advantages of Suno v6
- Ultra-fast generation: After entering a prompt, a finished product can be produced in seconds to about 30 seconds, with mass-production efficiency significantly higher than similar products.
- High degree of complete-song completion: Generate a complete song of up to 4 minutes in one pass, with coherent structure and optimal lyric rhythm, without the need for multiple extensions and stitching.
- Precise and controllable: Supports natural language local editing, word-level lyric modification, and sample extraction, bringing the granularity of creative control to an industry-leading level.
- Fully licensed training: Trained from scratch based on licensed music libraries from Warner, BMG, and Believe, with legal risk far lower than competitors.
- Open ecosystem: Works can be downloaded and distributed via Believe/TuneCore to third-party platforms such as Spotify.
- Layered model suite: The flagship v6, experimental v6-wild, and free v6-mini cover everyone from professional producers to ordinary enthusiasts.
Project address of Suno v6
- Project official website：https://suno.com/blog/introducing-v6
Comparison of Suno v6 with similar competitors
| Dimension | Suno v6 | Udio |
|---|---|---|
| Generation speed | About 30 seconds to generate 4 songs at the same time | About 45 seconds to generate 2 songs |
| Complete song duration | Generate up to 4 minutes at a time | About 2 minutes per generation, requiring multiple extensions |
| Distribution mode | Open: downloadable and distributed via Believe/TuneCore to third-party platforms | Limited to playback within the platform and sharing within the ecosystem |
| Product strategy | A three-tier model suite for all audiences + iMessage social distribution | A professional-oriented workflow focused on studio-grade mixing and stem output |
| Free availability | v6-mini is fully free | Limited free quota |
Application scenarios of Suno v6
- Independent musician demo creation: Songwriters enter lyrics and style descriptions and produce a complete commercially distributable demo in seconds, greatly shortening the creative validation cycle.
- Short video/self-media soundtrack: Creators upload videos or images as references to generate soundtracks, avoiding copyright risks while matching the emotional tone of the visuals.
- Advertising and brand music: Marketing teams generate multiple versions of advertising songs in batches according to product tone and directly distribute and launch them via Believe/TuneCore.
- Fan remix interaction: Use the upcoming artist-authorized remix feature to create remixes around idol works and receive revenue shares, activating the fan economy.
- Music education and inspiration: Music enthusiasts and beginners can explore unknown styles through v6-wild, or use natural language editing to polish their own works, lowering the threshold for creation.