Stop Guessing...
Start Creating the Song You Actually Want.
Built on a level of SUNO research nobody else has. 3,200+ verified tags, 16 genre families, 1,700+ sub-genres, 430+ instruments — organized into a smart 12-step system with live collision detection, character counter, line-level control, and quality score.
Everything free. No signup. No paywall.
Every category mapped in depth · all free · all searchable
SUNO on its own vs. SUNO with AceTagGen
Probability-based creation vs. intent-based production.
Without AceTagGen
You give SUNO a direction. The model averages thousands of similar songs and hands you the statistically safest path. Your intent becomes a rough hint — the AI picks the rest.
- ·The Slot Machine Effect. Without precise anchors, SUNO falls back to the median of its training data. You spin — you take what lands.
- ·Dynamic flatness. The chorus doesn't really explode. The verses don't really breathe. Without tension-and-release instructions, one energy level runs the whole track.
- ·Vocal drift. Verse 1 and verse 2 are close — not identical. Without an anchored vocal style, tonal consistency slips between sections.
- ·Intent lost in translation. You heard a specific song in your head. You got a pleasant song that's close. “Close” is a different song.
Like ordering food without seeing the menu. You'll get something edible — maybe not what you craved.
With AceTagGen
You stop guessing and start producing. Every axis SUNO actually responds to — surfaced, structured, guided. Before you click generate, your prompt is already the song you intended.
- ✓Sub-genre precision. 16 genre families → 1,700+ sub-genres pinpoint the exact sound cluster. Not “rock” — “90s grunge” vs “modern post-rock” hit completely different clusters in SUNO's model.
- ✓Dynamic control. Tension-and-release bracket tags ([Build-up], [Drop], [Breakdown], [Final Peak]) assigned line-by-line. Your chorus lands because you told it to.
- ✓Vocal anchoring. 180+ vocal styles locked at the Style level + reinforced per-line in the Song Editor. One voice across the whole track.
- ✓Intent → structure. The feeling in your head becomes BPM + key + mood + vocal delivery + tag order. SUNO stops guessing because you stopped asking it to.
- ✓Built on SUNO's actual rules: 1,000-char Style field enforced, per-category limits (Genre 1-2, Mood 1-2, Instruments 2-3), optimal tag ordering auto-applied.
- ✓Collision detection + quality score: catches conflicting tags before you waste credits. 100-point validator before you paste.
You're not commissioning a song. You're producing one — down to the single tag.
See the Difference
Same song idea. Two prompts. The precision is the difference.
“Pop song, upbeat, female vocals”Genre vague. Tempo implied, not set. Mood missing. Production clueless. SUNO picks from average of all “pop” — could be bubblegum, could be synth-pop, could be K-pop. Rolling the dice.
“Modern chart pop, bright, catchy, female vocal, 118 BPM, G major, modern pop polish”Sub-genre anchored first (weight priority). Two moods max (per-category limit). BPM numeric. Key specified. Production cue locked. SUNO pulls from a specific training cluster — predictable, repeatable, professional.
Built Different
A full 3:40 track I made with nothing but structured tags — to show what the system can do when every detail is intentional.
Proven Prompt Techniques
You Won't Find Elsewhere
Named, documented, reproducible techniques — distilled from across the SUNO research database and built directly into the system.
The Sandwich Method
Place your dominant instrument at the start AND end of the Style field. SUNO anchors to both positions and reinforces the sound throughout the track.
The Nicholson 7-Word Formula
Genre + Tempo + Mood + Instrument + Vocal + Era + Reference — the exact structure behind a prompt that generated 3M streams.
The 2-Minute Rule
Never generate 4 minutes at once. Build 1:30 first, use Extend at the chorus endpoint. Hit rate jumps dramatically — SUNO handles short contexts much better.
Phonetic Acting
Spell stutters: “I-I-I miss you”. Melisma: “lo-o-o-ve”. Laughter: “HAHA” in caps. SUNO performs the acoustic gesture exactly as written.
Hardware Emulation
“TR-909 kick” instead of “drum machine”. “Moog bass” instead of “synth bass”. Specific hardware tokens pull from higher-quality training clusters.
Strong Token Preference
“Tremolo strings” > “cinematic”. Concrete sound tokens beat abstract emotional words. The system nudges toward tokens SUNO actually responds to.
Bracket Sandwich
Structure tag + style modifier stacked on separate lines: [Verse] / [soft piano] / [intimate]. 2-3 bracket tags per section max — more overloads the prompt.
Genre-First Ordering
The first tag in Style gets ~2× the weight. Always lead with genre — never tempo, never mood. The system auto-orders so the DNA anchor is position 1.
Per-Section Placement
Assign each instrument to specific song sections (Piano → Intro, Guitar → Chorus). The difference between a flat prompt and a fully arranged track.
Built for a Global Community
Works in 12+ languages — prompts, lyrics, AI Chat. Research drawn from SUNO communities worldwide: US, Brazil, Japan, Germany, Spain, Korea, France, China, India, Mexico, Italy, and more.
For Developers — Free & Open
The core scoring algorithm is open-source under MIT. Integrate the 100-point prompt validator into your own tools — no auth, no rate limits, no cost.
- ▸
npm i suno-prompt-scorer - ▸Public REST API:
acetaggen.com/api/suno-score - ▸Gradio Space on HuggingFace — interactive demo
Every Rule SUNO Follows —
Mapped Into the System
Not tag-stuffing. Not guesswork. A system that knows how SUNO's engine actually weighs, ranks, and renders your prompt — and builds around those rules automatically.
Tag Weight Ordering
The first tag in your Style field carries roughly 2× the weight of the second. By position 6, you're at ~15% influence. The system auto-orders tags so the ones that shape the sound actually land at the top.
Bracket Tags = 10× Power
Bracket tags in the Lyrics field ([Chorus | whispered]) are up to 10× more powerful than the same tag in the Style field — one of the core mechanics SUNO follows. The Song Editor puts 300+ bracket tags at your fingertips, line-by-line.
Strong vs Weak Tokens
“TR-909 kick” pulls from a specific training cluster. “energetic” is vague and interpreted loosely. Every tag in the library is classified — the system nudges you toward concrete tokens SUNO actually responds to.
Effect Placement Intelligence
75+ sound effects split across Natural · Urban · Musical · Vocal — and they don't all go in the same place. Ambient effects go in the Style field. Line-triggered effects (stutters, screams, crescendos) go as brackets in Lyrics. The system places each effect in the right field automatically.
What the Scorer Actually Checks
Every prompt you finish runs through a 15-point validation grid built on the full research database. The score isn't magic — it's 15 specific, testable checks, each with a clear explanation.
Character Limit Compliance
V4.5+ Style field = 1,000 chars. Lyrics = 5,000. Prompt overflow is silently truncated by SUNO — the scorer warns before you paste.
Tag Weight Ordering
First tag carries ~2× weight. Position 6 drops to ~15%. The scorer validates your genre is anchored at position 0 (most common mistake).
Per-Category Limits
Genre 1-2 · Mood 1-2 · Instruments 2-3 · Vocal 1 · Production 1-2 · Tempo 1 · Era 0-1. Exceeds limit in any category → scored down.
Invalid Tag Detection
Flags 50+ known tags SUNO doesn't respond to (e.g., [Forte], [Head (Melody)], [Sforzando]). Suggests verified replacements.
Genre Collision Detection
Catches conflicting pairs: lo-fi + crisp, calm + aggressive, jazz + electronic. 100+ known collisions flagged with specific reasoning.
Strong vs Weak Tokens
Concrete tokens (TR-909 kick, Moog bass) pull from specific clusters. Vague tokens (epic, vibe) get interpreted loosely. Scorer rewards specificity.
Sub-Genre Specificity
“Rock” alone = average of all rock. “90s grunge” = specific cluster. The scorer nudges you from parent genres to specific sub-genres.
Bracket Syntax Validation
[] for directors (not sung) · () for performance (sung) · // for technical notes. Scorer catches misused syntax that breaks SUNO output.
Hardware Token Preference
“TR-909 kick” > “drum machine”. “Moog bass” > “synth bass”. Specific hardware tokens pull higher-quality training data — scorer recommends upgrades.
Vocal Delivery Validation
Checks vocal tags against 180+ verified deliveries. Flags contradictions: [whispered] + [powerful], [melismatic] + [monotone].
Version Awareness
Some tags work in V4.5 but break in V5 (and vice versa). Scorer detects version-specific issues and suggests appropriate alternatives.
Tag Density Check
Under 4 words = too sparse. Over 15 = diluted signal. Sweet spot of 5-8 precise tags consistently outperforms tag-stuffed prompts.
Bracket Tag Count per Section
Max 2-4 bracket tags per section. More = SUNO ignores most of them. Scorer counts per-section bracket density in Lyrics field.
Multi-Language Compatibility
Works on prompts in 12+ languages. Detects when language-specific tags (maqam, raga, sertanejo) need regional-specific pairing.
Final Weighted Score
Length 30% · Collisions 30% · Specificity 25% · Density 15% — weighted average normalized to 100 points with color-coded breakdown.
Free · no signup · instant feedback
Built for Musicians — Friendly for Everyone Else.
Every layer of music theory that actually affects SUNO output is mapped into the system. If you speak the language, you'll feel at home. If you don't — Auto mode has you covered at every step.
C · D · E · F · G · A · B — Major and Minor across all major keys. Plus Bb, Eb, Ab, F# for jazz and classical precision.
Dorian · Mixolydian · Lydian · Phrygian · Locrian — the modal vocabulary for jazz, film, world, and progressive music.
Pentatonic · Blues · Harmonic Minor · Melodic Minor · Chromatic · Whole Tone · Arabic · Major · Minor Keys.
I-V-vi-IV · vi-IV-I-V · I-vi-IV-V · ii-V-I · 12-bar Blues · chromatic transitions · modal interchange.
4/4 (common) · 3/4 (waltz) · 6/8 (compound) · 5/4 (Take Five) · 7/8 (odd meter). For anything beyond basic pop.
Straight · Swing · Shuffle · Half-time · Double-time · Syncopation · Rubato. Feel is encoded, not just tempo.
BPM-accurate presets across 44 tempo descriptors. Numeric BPM gives SUNO ~3× the tempo accuracy of words like “fast”.
Not a musician? Skip all of the above with one click. Auto mode picks genre-appropriate defaults you don't have to think about.
“I don't know music theory” isn't a barrier here — it's just a reason the system has defaults. The depth is there when you're ready for it.
Depth Without the Friction
A system this deep is only useful if it feels effortless. Every interaction was designed to keep you moving — from first click to finished prompt.
Audio Preview — Every Instrument
Click any of 430+ instruments to hear its actual sound before you pick it. No more guessing what a “Mellotron” or “Duduk” sounds like.
Research Hint on Every Step
Each of the 12 steps shows a one-line insight from the research — e.g., “Genre is the #1 tag — SUNO weighs it ~50%”. A mini SUNO course while you build.
Live Preview Panel
Color-coded prompt builds in real time as you move through steps. Character counter, category tags, and quality signals — always visible on the right.
Live Collision Warnings
Picked “Calm” + “Aggressive”? Two opposing production styles? The system flags it instantly with a one-click “Fix” button.
100-Point Quality Score
Every finished prompt gets an automatic score across length, tag collisions, specificity, and density — with specific improvement tips before you paste.
Auto-Save & Resume
Step out mid-build, come back days later — every choice is saved locally. A banner invites you to continue exactly where you left off.
Keyboard Shortcuts
Ctrl+Enter next · Ctrl+Backspace back · Esc undo. Fly through the flow.
Prompt Library
Save builds to your personal library. Rate, filter, remix, and reuse. Build a private collection of your best prompts.
One-Click Copy & Export
Copy the prompt directly · Open SUNO with one click · Export as Text, JSON, or PNG image for social sharing. Ship in seconds.
Every Tag Explained
Hover or tap any tag to see what it does, when to use it, and which genres it fits best. Learn the system as you build.
AI Lyrics Writer · 12 Languages
Need lyrics? Built-in AI writer generates them in 12 languages, matching your genre, mood, structure — with bracket tags auto-placed.
Auto Song-Structure Detection
Paste raw lyrics. The system splits them into sections automatically — Intro · Verse · Chorus · Bridge · Outro — and inserts the right bracket tags. No manual tagging for standard structures.
A Rich Blog of SUNO Secrets
Deep-dive articles, power tricks, bracket syntax, tag weighting, vocal techniques — the real mechanics of SUNO explained. Updated regularly with new findings.
Two Ways to Build Your Song
Use the structured 12-step system yourself (free), or let AceTagGen AI handle it (paid).
AceTagGen AI — An AI With a Producer's Mindset
Trained on the full research system — tag weighting rules, bracket power hierarchy, SUNO version-specific behaviors, genre-specific instrument mappings across 1,700+ sub-genres. Describe your song in plain language. Returns a production-grade Style + Lyrics + tag set, structured exactly how SUNO responds best.
- Knows SUNO's actual mechanics — not generic AI guessing
- Auto-applies per-category limits + tag ordering rules
- Explains its choices — learn as you use it
- English, Hebrew, Spanish, Portuguese, Japanese, and more
Section-Level Instrument Placement
One of SUNO's strongest response patterns: precision at the section level. Assigning instruments to specific song parts (Piano → Intro, Guitar → Chorus, Strings → Bridge) consistently produces more arranged, layered outputs than listing instruments generically in the Style field. The system builds this structure automatically — sub-genre-aware, per-section, editable.
- Per-section assignment: Intro · Verse · Chorus · Bridge · Outro
- 430+ instruments mapped across 1,700+ sub-genres
- Smart defaults suggest what fits — you override anytime
Song Editor — Line-by-Line Control
Bracket tags placed directly before a lyric line are dramatically more effective than the same tags in the Style field — up to 10× the impact. The Song Editor exposes this power: drag-drop sections, add emotion, voice changes, stutters, echoes, crowd vocals at the line where they matter. 300+ bracket tags, each with a short description of what it does and when to use it.
- Drag-drop song sections: Intro · Verse · Chorus · Bridge · Outro
- 300+ bracket tags — emotion, voice, structure, effects, dynamics
- Every tag explained — no guessing what [crescendo] or [808 kick] does
- AI Lyrics Writer built in · 12 languages
A Prompt You Can Copy, Save, and Reuse
Every build ends with a clean Style field, structured Lyrics, a Quality Score, and one-click copy straight into SUNO. Save your best builds. Export as text, JSON, or image. It's yours — all of it. No account required.
- Copy all · Open SUNO · Save to library
- Export: Text · JSON · PNG Image
- Quality Score with specific improvement tips
Help us ship the next wave
Deeper research articles. Tag updates for every new SUNO model. A ready-made prompt library for every genre family. A Chrome extension inside SUNO itself. All free for everyone — funded by coffee-sized tips and the AI Chat subscribers.
See how to supportThe deepest SUNO system. Completely free.
3,200+ verified tags. A smart 12-step builder. Line-level Song Editor. Quality score. Full research blog.
No signup. No paywall. No trial. Just start.
⚡90 seconds·No signup·Everything free