> For the complete documentation index, see [llms.txt](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/docs/suno/music_with_max_suno_v5_5_guide.md). # Music with Max Suno v5.5 Guide A practical, music-first roadmap for turning creative ideas into reliable Custom Mode prompts through the Music with Max platform layer. > This guide describes a third-party prompt-building layer that sits on top of Suno. The categories, limits, and labels below are written for that layer's workflow. They should not be read as a one-to-one map of Suno's native controls. ## Core Concept Treat every prompt like a mix: * The first style descriptors set the groove. * The mood colors the harmony and delivery. * The subject gives the song a story or setting. * The audience shapes arrangement choices. * Lyrics and metatags cue structure, performance, and transitions. * Excluded styles keep unwanted elements out of the prompt recipe. ## Prompt Categories | Category | Purpose | Where It Matters | | -------------- | ---------------------------------------------------------- | --------------------------------------- | | Tags | Genre, instrumentation, texture, era, and production style | Primary style direction | | Moods | Emotional atmosphere, intensity, and delivery | Style direction and section tone | | Subjects | Theme, scene, narrative, or creative world | Lyrical intent and optional style color | | Audiences | Listener, activity, or use context | Tempo, density, energy, and arrangement | | Lyrics | Words, song sections, and performance metatags | Structure and performance roadmap | | Exclude Styles | Sounds, genres, or traits to avoid | Negative prompting | Use Tags and Moods to shape the Style field. Use Subjects and Audiences to shape the creative intent. Use Lyrics and metatags to cue structure and performance. ## Music Tags Tags are descriptive markers that define the song's style DNA: genre, instrumentation, texture, scene, and era. Common tag types include: * Genre or style indicators, such as "Future Garage," "Darksynth," or "Lofi" * Instrumentation and production details, such as "analog arpeggios," "gated reverb drums," or "deep sub bass" * Functional purposes, such as "Study Music" or "Sleep Aid" * Cultural contexts, such as "Cyberpunk" or "Retro" * Atmospheric qualities, such as "Ethereal" or "Industrial" * Time or setting references, such as "Nocturnal" or "Urban" Tags are usually the strongest influence on the genre and style of the generated song. Front-load the most important descriptors because early terms tend to carry the most weight. Avoid artist-name prompts such as "sounds like Burial." They are unreliable, may be filtered, and are less useful than descriptive equivalents such as "deep sub bass, shuffled garage drums, rain-soaked atmosphere." Avoid negative tags such as "no drums"; unwanted elements belong in Exclude Styles. ## Music Moods Moods describe the emotional gravity behind the sound. They can influence melody, harmony, arrangement density, and vocal delivery. Useful mood types include: * Emotional states, such as "Melancholic" or "Triumphant" * Energy levels, such as "Intense" or "Calm" * Mental states, such as "Focused" or "Dreamy" * Atmospheric feelings, such as "Dark" or "Ethereal" Place moods after the main genre and instrumentation descriptors. For example: ``` Synthwave, darksynth, analog arpeggios, gated reverb drums, nostalgic, energetic, dark ``` Mood can also be applied inside the lyrics through short section tags: ``` [Bridge: somber, stripped back] ``` ## Music Subjects Subjects define what the song is about or where it lives conceptually. Common subject types include: * Narrative themes, such as "mythic journeys" or "redemption" * Settings, such as "corporate dystopia" or "desert landscapes" * Activities, such as "getting to sleep" or "high-energy study" * Cultural references, such as "8-bit gaming" or "urban life" * Personal experiences, such as "inner struggle" or "personal growth" Subjects primarily shape lyrical direction. They can also add one or two useful style colors, such as "dystopian," "desert blues," or "neon nightscape." ## Music Audiences Audiences describe who the track is serving and why. In this platform layer, audience context helps translate listener needs into musical choices. Useful audience types include: * Activity context, such as "students cramming" or "remote workers" * Cultural identity, such as "80s kids" or "cyberpunk enthusiasts" * Professional background, such as "software developers" or "creative professionals" * Emotional needs, such as "mental wellness seekers" or "mindful listeners" * Lifestyle context, such as "digital nomads" or "cafe dwellers" Audience choices should affect practical production decisions. A track for "students cramming" suggests steady tempo, low vocal distraction, and controlled dynamics. A track for "cyberpunk enthusiasts" suggests darker synth textures, harder edges, and more dramatic sound design. ## Music Lyrics Lyrics include both the words to be performed and the structural or performance direction that shapes the song. Use bracketed metatags at section boundaries. Common metatag types include: * Structural tags: `[Intro]`, `[Verse]`, `[Pre-Chorus]`, `[Chorus]`, `[Bridge]`, `[Breakdown]`, `[Drop]`, `[Hook]`, `[Interlude]`, `[Instrumental Break]`, `[Guitar Solo]`, `[Outro]`, `[End]`, `[Fade Out]` * Parameterized section direction: `[Verse: whispered vocals, acoustic guitar only]` or `[Chorus: anthemic, full band, layered harmonies]` * Vocal and performance tags: `[Female Vocal]`, `[Male Vocal]`, `[Spoken Word]`, `[Whispered]`, `[Choir]`, `[Harmonies]`, `[Ad-lib]`, `[Vocoder]` * Texture and arrangement cues: `[Intro: vinyl crackle fades in, warm jazz piano]` or `[Outro: elements fade, long reverb tail]` * Instrumental sections: `[Instrumental]` Use the Instrumental toggle for a fully vocal-free track. It is more reliable than relying on lyric tags alone. ### Lyrics Formatting Rules Bracketed tags guide the model; they do not command it. Everything in brackets is soft conditioning that the model tries to match, not a guaranteed instruction. Recognized musical vocabulary works well: "reverb," "filter sweep," "fade," "swing," and approximate BPM values can nudge the output. Exact numeric precision is not reliable. Prompts such as "30% wet," "12kHz," or exact BPM values do not map to measurable production controls. Avoid timestamps such as `[0:23-0:45]`. The model follows section sequence, not clock positions. Keep metatags short: one to three words, or one short descriptive phrase. Overly long or unfamiliar tags can be misread or sung as lyrics. Describe the sound rather than the engineering setup: * Better: `warm pad swells in, heavy reverb` * Riskier: `30% wet reverb at 0:23` Sample-source references such as "grandma's voice sample" or "Blue Note LP" are treated as vibe descriptions only. The model cannot pull real samples from text. To incorporate real audio, use the platform's available upload, cover, or studio workflow. Aim for roughly 30-40 lyric lines with section tags. Very short lyric inputs, especially under about 15 lines, tend to produce short songs. ## Producing Music The Lyrics category can be as long as needed within the platform's working limit, which is roughly 5,000 characters in this guide's workflow. For Tags, Moods, Subjects, and Audiences, use comma-separated lists with a practical limit of 250 characters per category. Tags and Moods together should fit comfortably inside the Style field. A practical sweet spot is 4-7 total style descriptors. Fewer than 4 descriptors can leave too much to chance; more than about 7 can make descriptors compete and produce unfocused output. Always put the most important terms first because long fields may be truncated or diluted. ### Field Reference | Field | Practical Limit | Notes | | -------------- | ------------------------ | ---------------------------------------------------------------------------------- | | Style of Music | 1,000 characters | Built from Tags, Moods, and optional Subject color; front-load the strongest terms | | Lyrics | Roughly 5,000 characters | Lyrics plus metatags | | Title | 80 characters | Cosmetic; do not rely on it to shape the music | | Exclude Styles | Comma-separated list | Best place for negative direction, such as "EDM, autotune, trap hi-hats" | ### Creative Sliders * Weirdness: Low values are more conventional; high values are more experimental. * Style Influence: Higher values follow the Style field more closely. Use higher values when the prompt is precise. * Audio Influence: Controls how strongly uploaded or reference audio shapes the result. ### Personalization Features * Voices: Record or upload a singing voice and use it as the vocalist in generations, subject to the platform's verification and plan requirements. * Custom Models: Train private model variants on stylistically consistent tracks from your own catalog, subject to the platform's plan requirements. * My Taste: Lets the platform learn recurring genres and moods and bias suggestions toward them. A detailed Style prompt should take priority over this default bias. ### Other Capabilities Depending on plan and platform availability, the workflow may include longer generations, multiple variations per create action, song editing or section replacement, stem separation, studio-style editing, and prompt enhancement. Use prompt enhancement when learning descriptor vocabulary. Disable it when you need precise, repeatable production control. ## Recommended Generation Loop 1. Ideate with several variations that use different Style descriptors. 2. Select the best one or two outputs. 3. Refine descriptors and metatags. 4. Extend the strongest result. 5. Edit weak sections. 6. Export once the arrangement is stable. The same prompt can produce different results each time. Try a promising prompt two or three times before changing it. ## Complete Example This example shows how the platform-layer categories combine into a Suno-facing prompt recipe. **Tags:** Synthwave, Darksynth, Retro, driving mid-tempo, analog arpeggios, gated reverb drums, 80s production **Moods:** Nostalgic, Energetic, Bold, Dark **Subjects:** Digital Age, Night Driving, Urban Life **Audiences:** Retro Gaming Fans, Night Shift Workers, Cyberpunk Enthusiasts ### Resulting Style Field ``` Synthwave, darksynth, driving mid-tempo, analog arpeggios and gated reverb drums, retro 80s production, nostalgic, energetic, dark ``` ### Exclude Styles ``` acoustic, country, ballad ``` ### Lyrics Structure ``` [Intro: pulsing arpeggio synth, rain ambience] [Verse 1] Neon lights through city rain Digital dreams in my brain Racing through these streets again [Pre-Chorus: drums build, thunderstorm ambience] Chrome and steel reflect the night Memories fade in binary light [Chorus: anthemic, layered synths] Data streams into the sky Breaking through the firewall somehow Time is running out tonight [Bridge: all elements drop except pulsing bass and rain, new pad swells in] Back into the mainframe now System overload somehow [Chorus: vocoder harmonies, building intensity] Racing through these streets again Neon lights through city rain [Outro: full arrangement, heavy saturation, long fade] ``` For this example, set Style Influence high because the prompt is specific. Set Weirdness low-to-medium unless you want a more experimental result. ## Final Note Start simple, generate multiple takes, keep what works, and refine one variable at a time. Strong Suno prompting is less about one perfect prompt and more about a repeatable creative loop. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/docs/suno/music_with_max_suno_v5_5_guide.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.