Loudly is a music workspace for original tracks, remixes of songs you own, mastering, stem splits, and a release step. Generation is split across instrumental VEGA models, vocal model MANTA 1, and partner sample model Stability 2.5. SongDNA stores a fingerprint for later prompts and captions. The Music API adds parameters, a catalog, playlists, stems, and lossless WAV.
Official website · www.loudly.com
Use Loudly when you need the model menu, not one generate button. VEGA stays instrumental. MANTA 1 is the vocal model. Stability 2.5 is the partner sample path. SongDNA is a fingerprint, not another generator, and the Music API is a separate key-and-volume job. The homepage says proprietary output is cleared at generation. Terms dated 21.01.2026 say Loudly does not warrant that output is non-infringing or commercially ready without more clearance, and Partner Output is not licensed by Loudly. Distribution headings mention 50+ platforms and keeping royalties. The terms say Loudly only facilitates delivery and does not guarantee acceptance or a payout.
The homepage says you use various models to generate original tracks, remix songs, and add FX and mastering in one workflow. You choose proprietary or partner models by speed, quality, cost, or sonic style. One platform, multiple generation paths, more creative control. VEGA 1, the low-latency instrumental, is raw, fast generation. Ideal for scratch tracks and rapid iteration when speed matters more than polish. VEGA 2 offers precise control over instrumentation, arrangement, and vibe for background music, sync, and client delivery. VEGA 3 is the highest-capability VEGA model for finished instrumental output. MANTA 1 is full song generation with lyrics and vocals. The only model in the stack for complete vocal track output. Stability 2.5 is the partner sample model. The page says proprietary output is cleared at generation. The terms narrow that claim.
Official documentationSongDNA analyzes audio, lyrics, structure, mood, tempo, key, hooks and your own creative inputs to build a usable fingerprint of any track. Loudly calls that fingerprint a context window for playlist pitches, captions, cover art, and video prompts. Diagram inputs are audio analysis, lyrical analysis, and optional user inputs. The shown example shows A minor, 124 BPM, high energy, a tense mood, and a hip-hop, trap, and R&B split. Named outputs are a style prompt, a Spotify pitch, captions, and an on-request short-form cut. The page says agent mix, arrangement, and FX choices are grounded in SongDNA, and that there is no intelligent agency without it. Was not verified here.
Official documentationRemix copy says the engine does not create soundalike covers. It analyzes songs you own, separates elements, and rebuilds versions. Mastering is separate. Upload your track, choose a mastering style and export a polished version guided by platform loudness targets. Stems are another control. Split any track into clean, DAW-ready stems. Separate vocals, drums, bass, guitar, synth, strings and more in high-quality WAV or MP3. The heading says release to 50+ platforms and keep 100% of royalties. In the terms, royalty-free means Loudly does not charge ongoing royalties for licensed Loudly Output, not that other claims disappear. Loudly acts as a technical facilitator of the distribution workflow and does not guarantee acceptance, continued availability, monetization, or specific performance outcomes for any release.
Official documentationAPI blocks are concrete. Text-to-music takes a prompt. The generator takes parameters such as genre, energy, and duration. The catalog is 3,500 tracks ready-to-go. Playlists auto-generate 50+ songs. AI radio is a non-stop stream. One block offers 500+ expert prompts. Stems retrieve isolated audio, sample packs retrieve samples, and studio output is lossless.WAV files. The FAQ says customize audio by genre, tempo, energy, and duration, on one endpoint. It also cites 1000+ pre-set prompts, which disagrees with the 500+ line above. You will automatically receive an API key with a free track allowance to begin integrating AI-driven music features into your workflow after a developer-portal signup. Price follows business size, monthly volume, and Basic versus Premium. Price per track decreases the more you use. A dollar rate was not verified here.
Official documentationFor a short bed starts on VEGA 1, the fast instrumental, then moves to VEGA 2 for a client cue or VEGA 3 for the finished instrumental path. Keep the first prompt small enough that you can tell whether the lane is right. The homepage capture states no maximum length. Duration appears as a Music API parameter. Stems mention WAV or MP3, and the API mentions lossless WAV. Check the file you download before you cut picture to it.
A suggested catalog pass runs SongDNA before a remix. Upload audio you own, read the fingerprint, then request a style prompt or a playlist pitch. The on-page sample shows key, BPM, mood, chords, and a genre split. Those figures are the illustration, not a reading of your file. Remix copy limits sources to songs you own and rejects soundalike covers. Master next if you need a louder export, or split stems if you need parts. Do not treat the heading about keeping 100% of royalties as the payout rule. The terms allow deductions and do not guarantee monetization.
A suggested integration starts on the developer portal. Take the key with a free track allowance, then choose a prompt, parameters, or one of the 3,500 catalog tracks. Playlists are described as 50+ songs. Set genre, tempo, energy, and duration when the cue must hit a length. No call was made for this profile. Stems and sample packs fit an app that needs parts. License type is Basic versus Premium, and the per-track price is said to fall as volume rises, with no number public. Partner Output is licensed by the partner, not by Loudly.
Open the generator. Use a VEGA model for an instrumental, MANTA 1 for lyrics and vocals, or Stability 2.5 for samples.
Instruction inputs are text prompts and selections such as genre. The other accepted upload is audio. Other file types are refused unless a feature expressly enables them.
Use SongDNA for a fingerprint, pick a mastering style after upload, or split the track into WAV or MP3 stems.
Loudly Output sits under a separate license agreement. Distribution is a partner workflow. API audio has its own commercial-coverage wording.
Terms say the service is available in both free and paid tiers. No dollar amount appears in the public quotes, so the prices below are Not verified.
Trying the stated model stack before a paid tier.
Not verified.
Free tier; numeric limits were not in the usable capture
Use past the limits described for the free tier.
Not verified.
Paid tier, named only as a class of plan in the terms.
The pricing URL returned only a title, so third-party dollar figures are not copied here. API copy says the per-track price falls with use and mentions pay-as-you-go, without a number. A free API allowance is not the same sentence as the website free tier.
You want the entertainment-tagged song product instead of this model menu and API.
Loudly's pages instead center on VEGA, MANTA, SongDNA, and a Music API.
You are comparing another Audio & Music generator without Loudly's API.
Soundraw is a separate Audio & Music entry. Consider it when you do not need Loudly's VEGA, MANTA, and developer-key split.
Choose Soundful when you need video as well as audio, not only Loudly’s SongDNA, stems, and Music API.
Soundful also carries an AI Video tag. Loudly's pages stay on generation, SongDNA, stems, and the Music API.
You want another catalog audio product rather than Loudly's studio plus API.
Loudly adds a partner-model split, API stems, and a distribution service it only facilitates.
The terms say the service is available in both free and paid tiers. The free tier includes certain features, with usage limits as described on the website. The homepage says to try all models free, with no credit card required.
SourceInput is text prompts and parameter selections, with genre as the example, plus audio you upload. Other file types are refused unless a feature expressly enables them. Output may be VEGA audio, partner output, or catalog audio, and a file available now may not stay available.
SourceVEGA 1, VEGA 2, and VEGA 3 are marked Instrumental. MANTA 1 is the vocal model. Stability 2.5 is the partner sample model.
SourceThe homepage says release to 50+ platforms and keep 100% of royalties. The terms say Loudly does not guarantee acceptance, availability, monetization, or a performance outcome.
SourceA developer account receives an API key with a free track allowance for prompts, stems, and catalog fetches. Paid use is volume-based, by business size and Basic versus Premium. A dollar rate was not verified here.
SourceThey license Loudly to process instruction inputs to run the service. The terms still refuse a warranty that output is non-infringing or commercially ready without more clearance.
SourceOct 6, 2026 · NavHunt editorial
NavHunt did not run Loudly, generate a track, upload audio, call the Music API, or submit a release. This profile restates the homepage, Music API page, and terms reviewed on 6 October 2026. The pricing URL returned a title only. Clearance and royalty lines are Loudly's claims. This description is based on Loudly's official materials reviewed on 2026-10-06, including Loudly homepage, Loudly Music API, Loudly terms and conditions.
Pick VEGA, MANTA, or the partner sample model, then check the license before a release.