Harmony vs MediaSage
Harmony
felipenipper.comMediaSage
github.comSide by side
| Figure | Harmony | MediaSage |
|---|---|---|
| Price from | Free | — |
| Free plan | Yes | — |
| Free trial | — | — |
| Open source | — | — |
| Web | — | Yes |
| Windows | — | Yes |
| macOS | — | Yes |
| Linux | — | Yes |
| iOS | Yes | — |
| API | — | Yes |
| Self-hosted | — | Yes |
| Plans published | 1 | 0 |
| Spec score · AI Music Playlist Generators | 8.5 (#6) | 7.4 (#19) |
| Maker | Felipe Nipper says he independently designs, develops, and ships his apps, and works as an iOS engineer. | The repository owner is identified on GitHub as Eric Wilson; no headquarters or founding date is stated on the maker pages opened. |
| Purpose | Harmony generates Apple Music playlists from descriptions of moods, moments, artists, genres, or other preferences. | MediaSage is a self-hosted web app that generates Plex playlists and album recommendations using music in the user’s library. |
Plans
In their own words
Harmony
Harmony is an iOS app that generates playlists from descriptions of moods, moments, activities, vibes, artists, genres, and other listening preferences. Describe what you want in plain language, then tap for another set of playlists shaped by your taste. Generated playlists save directly to your Apple Music library. An Apple Music membership is required, while Harmony itself is listed at free on the App Store. The maker lists SwiftUI, MusicKit, generative AI, and StoreKit among its technologies. Harmony was founded in 2024 and is offered for iOS. Its privacy policy says analytics cover usage frequency, time spent, and bugs or crashes. Information entered into the app, including music preferences and activity, is subject to the OpenAI Privacy Policy through its OpenAI services integration. The app is not intended for anyone under 13. Support is available at [email protected].
MediaSage
MediaSage is a free, MIT-licensed self-hosted web app for building Plex playlists and recommending albums from a music library. Users can describe a playlist in natural language or start with a selected song; prompts can include mood, era, instrumentation, genre, production style, and a seed track. Album suggestions can focus on owned albums or look beyond the library. Filters cover library genres, decades, minimum rating, and excluding live recordings. The app syncs Plex music data to a local SQLite database for filtering and track selection. Users can preview tracks, create or update playlists, and send tracks to active Plex clients. It works with Google Gemini, Anthropic Claude, OpenAI GPT, Ollama, and custom OpenAI-compatible APIs, and displays estimated provider costs before generation. Setup requires Docker or Python 3.11 or newer, a Plex server containing music, a Plex token, and an AI provider key or local model. The maker documents Linux, macOS, and Windows use, with NAS instructions for Synology, Unraid, and TrueNAS SCALE. Approximate provider capacity limits are listed, and local model support is described as experimental.