mpv-android: Open Source GPL Media App on F-Droid
mpv-android review: where mpv.conf and input.conf fit, why desktop mpv scripts and shaders aren’t promised, and Android TV support remains unverified.
App summary
- App
- mpv-android
- Developer
- mpv-android project contributors
- License
- GPL-2.0-or-later
- Version
- unknown
- Size
- unknown
- Price
- Free
Where mpv-android’s mpv.conf and input.conf fit—and what the fact data does not confirm
Anyone who has spent time with mpv on a desktop eventually goes looking for mpv.conf. It is where you pin down defaults, force a specific audio output, set subtitle styling, or override a decoder that misbehaves. Because mpv-android is described in the fact data as an Android frontend for the mpv media player built on libmpv, it is reasonable to assume that some of that configuration culture carries over. What the fact data does not do is confirm where those files live on Android, whether both are read at all, or what syntax the app accepts. That distinction matters more than it sounds. A frontend is not the same thing as the engine underneath it, and the storage model on Android is not the storage model on a Linux desktop. If you want to compare mpv-android’s playback approach with another free video player that emphasizes broad format support, the VLC for Android review covers its network streaming and subtitle handling. For a lighter-weight alternative that focuses on straightforward local video playback, the Just Player review explains its passthrough support and limited extras. Readers weighing mpv-android against a more feature-heavy player can see how MX Player review balances codec support, subtitles, and advertising.
This is not a minor documentation nit. The same fact data flags the app’s interface and settings as less beginner-friendly than mainstream players. When a player ships a settings surface that assumes some familiarity, config-file editing stops being an exotic corner case and becomes one of the realistic ways people get the behavior they want. So the honest answer to “where do I put my mpv.conf” is: the fact data does not say, and any guide that hands you an absolute path like /sdcard/Android/data/... is guessing unless it points back to the project itself.
The practical instruction, then, is to treat exact config paths as unknown and go to the source. The official repository at https://github.com/mpv-android/mpv-android is the place where current instructions would live, and the F-Droid listing is the other legitimate reference point for what a given build actually ships. If a blog post or a forum reply gives you a path with no date and no link to either of those, treat it as unverified.
One more thing worth stating plainly, because it shapes how you should think about backups and portability: mpv-android requires no account or sign-in. There is no profile to sync, no cloud store holding your preferences, and no server-side copy of your config to restore onto a new phone. Whatever configuration you build lives on the device, and moving it means moving it yourself. That is consistent with an open-source player that has no account system, but it also means config guidance should never imply that signing in will bring your settings back.
Desktop mpv scripts and shaders on mpv-android: compatibility is not a promised feature
The most common misconception about this app is that it is desktop mpv with a touchscreen bolted on. It is not. The fact data describes it as a libmpv-based frontend, which is a meaningful qualifier: libmpv is the playback library, and a frontend is the shell around it. The shell decides which knobs are exposed, which file paths are reachable, and which extension mechanisms are wired up. Broad codec and container support comes from the libmpv layer and is listed as a genuine strength. Lua scripting and user shader loading are separate questions entirely, and the fact data does not confirm either one on Android.
That gap is easy to paper over with wishful thinking. Desktop mpv users have accumulated a large ecosystem of scripts for things like automatic subtitle downloading, playback history, and quality-of-life keybindings, plus shader packs for scaling and color work. It is tempting to assume all of that transfers. Nothing in the fact data supports that assumption, and there is a structural reason to be cautious: shaders and scripts often assume desktop file layouts, desktop GPU paths, and desktop input models. A mobile frontend may expose none, some, or all of that, and the fact data does not tell us which.
Advanced users will still want to poke at interface and config customization, and that is a legitimate thing to do with an open-source app. Just go in with the right expectation. The app is already flagged as less beginner-friendly than mainstream players, so the advanced path is not a hidden bonus mode — it is closer to the main road for anyone who wants non-default behavior. Expect to read, not to be guided.
The responsible way to check is to look at the official repository rather than a listicle. Do not take any specific script or shader as confirmed working on Android until the project itself says so. If you find a claim elsewhere, check whether it links back to https://github.com/mpv-android/mpv-android and whether it is current. The fact data also leaves the latest version and release date unknown, which means even a correct claim from a year ago may no longer describe the build you install.
mpv-android on Android TV: remote navigation, leanback, and passthrough remain unverified
The fact data lists the platform as Android. That is the whole of it. It does not confirm Android TV remote-control support, it does not confirm leanback launcher integration, and it does not confirm AC3 or DTS audio passthrough. Those are three separate questions, and conflating them is how people end up disappointed with a set-top box install. A player can run perfectly well on a TV screen with a mouse or an air-mouse and still be miserable with a directional pad, because gesture-based controls and a configurable interface are not the same thing as a remote-first design.
The interface point deserves emphasis. mpv-android’s controls are described as gesture-based, and its settings as configurable but less beginner-friendly than mainstream players. Neither of those facts is evidence of TV-first design. Gestures assume a touch surface. A D-pad assumes focus traversal, visible focus states, and a predictable back behavior. Those are different engineering problems, and the fact data does not tell us whether the second one was solved.
Installation and updates add another wrinkle for TV owners. The verified facts put the app on F-Droid and host its source on GitHub, while Google Play listing status is explicitly unknown. On a phone, sideloading or using F-Droid is a minor inconvenience. On a television, where the browser is awkward and file transfer is often a chore, it changes the practical experience of keeping the app current. The fact data also notes that update cadence and availability depend on the distribution channel, which is exactly the kind of thing that bites on a device you do not pick up every day.
So treat Android TV use as device-testing territory. If you have a box or a TV stick and you want to try it, try it — the app is free and requires no account, so the cost of finding out is your time. But do not buy hardware on the assumption that remote navigation, leanback behavior, and AC3/DTS passthrough are all handled. If audio passthrough specifically is your reason for looking at this app, verify it on your own hardware against the official repository’s current documentation before you commit.
Why mpv-android is not an MX Player-style media library—and who should skip it
The fact data is unusually direct here: mpv-android is not designed as a media library or catalog app. There is no ambiguity to resolve. If what you want is a wall of thumbnails, recently played rows, folder scanning, and a search that understands your collection, this is the wrong tool, and no amount of configuration will turn it into the right one. It plays files. It does not curate them.
That single design decision filters the audience hard. Casual users who want a polished library and discovery experience should look at VLC for Android or MX Player instead. Both are named in the fact data as alternatives, and the stated reasons are telling: VLC for its broad format support and more mainstream interface, MX Player for its media library and playback features. Those are not consolation prizes. They are different products solving a different problem, and for most people who just want to tap a poster and press play, they are the better answer.
The second filter is patience with settings. The interface and settings can be less beginner-friendly than mainstream players, which is a polite way of saying that some playback problems are solved by understanding the app rather than by tapping a button labeled “fix it.” Non-technical users who are unwilling to edit config files for advanced playback should take that warning seriously. It is not a knock on the app; it is a statement about who it was built for.
Finally, do not position this as a mainstream streaming replacement. The fact data does not promise DRM playback for services like Netflix or similar platforms, and there is no catalog, no account system, and no sign-in. Anyone expecting to log into a subscription service and stream protected content is looking at the wrong category of software entirely. Similarly, if you need built-in Chromecast or DLNA casting controls, or video editing and transcoding features, none of those appear in the fact data and none should be assumed.
What the verified facts do establish about the app
Strip away the assumptions and a short, solid list remains. The Android package name is is.xyz.mpv. The license is GPL-2.0-or-later. The source code is hosted at https://github.com/mpv-android/mpv-android, and the app is available on F-Droid. It is free. It requires no account or sign-in. Those are the verified facts, and they are enough to make a first judgment: this is an open-source, auditable player distributed through channels that do not require a store account.
The license is worth a sentence of its own. GPL-2.0-or-later means the source is not just visible but carries obligations that keep derivative work open. For a media player, that matters to a specific kind of user — someone who wants to know what the binary does, or who wants to build it themselves, or who does not want to depend on a vendor’s continued goodwill. The fact data lists “open-source and auditable” as a strength, and that is the concrete mechanism behind it.
There is also a list of things the fact data explicitly does not know, and it is longer than you might expect. The exact latest version and release date are unknown. The APK size for the current release is unknown. The minimum Android version for the current release is unknown. Google Play listing status is unknown. That is not a flaw in the app; it is a limit on what any review can honestly assert. If you need to know whether the build will install on an old device, or how much storage it will take, the answer has to come from the F-Droid listing or the repository, not from an article.
What you can say with confidence is that the app is free, has no account layer, and is published under a copyleft license with a public repository. For a certain reader, that combination is the entire pitch. For everyone else, the unknowns above are the first things to check before installing.
The libmpv foundation, and why “plays everything” is a starting point rather than a promise
mpv-android’s core function, per the fact data, is playing local and network video and audio files using libmpv, with the app acting as an Android frontend for the mpv media player. The strength listed is broad codec and container support inherited from that foundation. This is a real advantage and it is also a bounded one. Broad support means the engine knows how to decode a wide range of formats; it does not mean every file you own will play without adjustment, and it does not mean the frontend exposes every engine option.
The feature list adds specifics that are worth taking at face value: hardware and software decoding options, gesture-based controls with a configurable player interface, and support for subtitles, audio tracks, and network streams. Those are the working parts. Note the phrasing on decoding — “options” — which is a different claim from “hardware decoding works everywhere.” Note also that network streams are supported and that the network permission exists for streaming, which is the piece of the privacy picture the fact data actually confirms.
Subtitles and audio tracks are listed as supported, which answers the basic question of whether the app handles them at all. It does not answer the follow-up questions. If subtitles are not showing or are out of sync, the fact data gives no confirmation of a subtitle delay control, a font override, or an encoding fallback. The same goes for playing SMB, WebDAV, or FTP shares without a separate file manager: network streams are supported, but specific protocol support is not confirmed anywhere in the fact data. Do not assume a protocol picker exists.
That is the honest shape of the app’s capability story. libmpv gives it a strong engine and a wide format net. The frontend decides how much of that you can reach from a phone screen, and the fact data tells us the frontend is configurable but not beginner-friendly. The gap between those two statements is where most of the real user experience lives.
Hardware decoding as a switch you may have to turn off
Hardware decoding is listed as an option, alongside software decoding. That pairing is the tell. If hardware decoding were universally reliable, software decoding would not need to be offered as a peer. The fact data does not claim that hardware decoding always works smoothly on every Android device, and the misconception that it does is one of the more expensive ones a new user can carry, because the symptom — a black screen, stuttering, or a file that refuses to start — looks like a broken app rather than a decoder mismatch.
The useful mental model is that hardware decoding is a negotiation between the file, the chip, and the driver, and the app is only one party to it. When the negotiation fails, switching to software decoding is the standard fallback, and the fact data confirms that option exists. It is not a magic fix for everything, and it will cost you battery and CPU headroom on high-bitrate content, but it is the first thing to try when a file that should play does not.
Where this gets genuinely hard is that the fact data does not enumerate which devices or codecs are affected. There is no compatibility list here, no known-issues page quoted, and no minimum Android version confirmed for the current release. So the guidance has to stay general: if playback misbehaves, treat decoding mode as a variable rather than a constant, and check the repository for anything current about your hardware. Anyone who tells you a specific device is fine is speaking from their own sample of one.
The same caution applies to the settings surface generally. The app’s interface and settings are flagged as less beginner-friendly than mainstream players, and decoder selection is exactly the kind of setting that lives behind that warning. Expect to find it, expect to have to think about it, and do not expect the app to diagnose the problem for you.
Distribution, updates, and the unknowns around version and size
The verified distribution facts are short: available on F-Droid, source at https://github.com/mpv-android/mpv-android, Google Play listing status unknown. That last item is doing a lot of work. It means a reader cannot be told to expect store-based updates, and it means any article that describes the Play listing is describing something the fact data cannot confirm. The honest framing is that F-Droid and the project repository are the reference points, and everything else is uncertain.
This connects directly to a listed weakness: update cadence and availability depend on the distribution channel. That is a structural property of open-source Android distribution, not a criticism of this particular project. F-Droid builds on its own schedule. Repository builds depend on whoever is building them. If you are the kind of user who wants a predictable monthly update, that expectation does not fit here, and you should know it before you install rather than after.
The unknowns compound. Exact latest version and release date: unknown. APK size for the current release: unknown. Minimum Android version for the current release: unknown. Those three numbers are the first things most people check before installing anything, and this review cannot supply them. What it can do is tell you where to look — the F-Droid listing for the packaged build, the repository for source and release notes — and warn you that any figure quoted elsewhere without a link to one of those is probably stale.
None of this is a reason to avoid the app. It is a reason to treat it as software you maintain rather than software a store maintains for you. For a GPL-licensed player with no account system, that is a coherent trade: you give up some convenience and you get a build you can inspect and a channel you can choose.
Verdict
mpv-android is a configurable libmpv frontend for people who value open-source playback and are willing to manage settings. That is the thesis, and the fact data supports it without much stretching. The engine is strong, the license is copyleft, the repository is public, the app is free, and no account stands between you and a video file. If those are the things you care about, the app is aimed at you.
It is not a polished media-library replacement for MX Player. The fact data says so directly: no library, no catalog, and an interface that is less beginner-friendly than the mainstream options. Casual users who want thumbnails, discovery, and a settings screen that explains itself should use VLC for Android or MX Player, both named as alternatives for exactly those reasons. Users who want built-in casting controls, video editing, transcoding, or guaranteed DRM playback for subscription services should look elsewhere entirely — none of those appear in the fact data.
It is also not a guaranteed Android TV or DRM solution. Remote navigation, leanback behavior, and AC3/DTS passthrough are all unverified. Config file locations are unverified. Lua script and shader compatibility is unverified. The latest version, APK size, minimum Android version, and Play Store status are all unknown. A review that filled those blanks in would be inventing things.
What remains is a specific, defensible recommendation. Install it if you want an auditable, GPL-licensed player built on libmpv, you are comfortable treating decoding mode and configuration as things you adjust rather than things the app fixes for you, and you are willing to get your builds from F-Droid or the official repository at https://github.com/mpv-android/mpv-android. Skip it if you want a library, a remote-first TV experience, or a player that never asks you to think about a config file. Both of those are reasonable positions. The app just happens to serve one of them.
Pros
- Open-source and auditable
- Based on mpv/libmpv with broad codec and container support
- No account or sign-in is required
Cons
- Interface and settings can be less beginner-friendly than mainstream players
- Not designed as a media library or catalog app
- Update cadence and availability depend on the distribution channel
The link goes straight to mpv-android project contributors. We do not host app files.
Frequently asked questions
Is mpv-android free?
The recorded pricing model is Free.
What license does it use?
It is listed under GPL-2.0-or-later.
Where can I download the official version?
The official link points to https://github.com/mpv-android/mpv-android.