Scryer vs. Sonarr and Radarr

Sonarr and Radarr are mature. If you run one of each, for one person, they work well.

Scryer is built for the setup most people end up with anyway: movies, series, and anime, a few people in the house, more than one quality tier, subtitles, lists, and requests. It does all of that in one process instead of five or six.

Moving an existing setup? Read the migration guide before you retire anything.

Where They Differ

Organizing media and people

	Scryer	Sonarr / Radarr
Scope	Movies, series, and anime in one catalog	One app per media type
Quality tiers	One library per tier in the same instance	One instance per tier (a 1080p Radarr and a 4K Radarr)
Users	Per-user permissions, scoped to libraries	One admin login, or a portal in front
Requests	Built in, approved per library	Seerr or similar
Lists	Per person. Trakt, Plex, Simkl, TMDb, MDBList, AniList, MAL, and custom URLs (beta)	Per instance
Settings	Inherited: global, then catalog, library, and title. Any level can override	Per instance, varied with profiles and tags
Sign-in	Forms, passkeys, MFA, or your media server account	Forms, or an external auth proxy
Metadata	Full TMDb and TVDB for every media type, in your languages, with cast	TVDB for series, TMDb for movies. Radarr can show movie info in one other language

Finding releases

	Scryer	Sonarr / Radarr
Indexers	One list, routed per facet. Can be shared back out as Newznab or Torznab (beta)	Prowlarr syncs a copy to each app
Parsing	Context-aware, matches on aliases and other languages	Regex, no context
Search coverage	Each item searched once per indexer, then RSS	RSS, plus the missing and cutoff searches you start yourself
Release preference	Personas plus Rego rules. Import your custom formats, subscribe to TRaSH or SeaDex packs	Custom formats, usually synced from TRaSH
Score visibility	Current score next to the score at grab time, with the breakdown per rule	Custom format score per file
Required audio	Required languages per library, checked against the file’s audio tracks at import	Radarr: one language per profile, read from the release name. Sonarr: custom formats
Movie availability	Announced, In Cinemas, or Released. Pick a region and a day offset, per facet, library, or title	Radarr minimum availability
Delay profiles	By tag, protocol, and library	By tag and protocol
Direct downloads	HTTP indexers and a built-in HTTP download client	None
Outbound network	Per indexer or client: HTTP proxy, Byparr, Trawl, WireGuard, or SSH tunnel	One HTTP or SOCKS proxy for the whole app. FlareSolverr through Prowlarr

Files and imports

	Scryer	Sonarr / Radarr
Library layout	Flat, or nested with folders by letter or year	Flat root folders
Archives in downloads	Extracted and imported, nested and split sets included	Import refused until something else extracts them
Recycle bin	Inside each library root, so a removal is a rename. Path and retention are yours	One folder per instance. On another volume, every removal is a copy
Scripts	On events, or on a schedule: interval, daily, weekly, cron	On events
Integrations	Plugins, including ones the community wrote	Compiled in
Packaging	A service, or a macOS menu bar app, or a Windows tray app with its own window	A service, or a Windows tray icon that opens your browser

Feature Coverage

Feature	Scryer	Sonarr / Radarr
Monitor titles, RSS grabs, interactive search	✅	✅
Completed download handling and manual import	✅	✅
Rename and organize files	✅	✅
Quality profiles, upgrades, delay profiles	✅	✅
Advanced release decision rules	✅	✅
Post-processing scripts and webhooks	✅	✅
Activity, history, wanted, cutoff unmet	✅	✅
Kodi-style `.nfo` metadata on import	✅	✅
Release push from autobrr	✅	✅
Import lists and watchlist sync	✅ beta	✅
Movie availability thresholds	✅	Radarr only
Multi-lingual metadata	✅	Radarr, one language
Cast and crew, including dub casts	✅	Radarr, no dub casts
Media requests with approval queue	✅	❌
Multiple libraries per media type	✅	❌
Per-user permissions	✅	❌
Passkeys and MFA	✅	❌
Anime as its own media type	✅	❌
Multi-season and complete-series packs	✅	❌
Required audio verified at import	✅	❌
Subtitle providers and sync	✅	❌
Nested library folders	✅	❌
Scheduled scripts	✅	❌
Native HTTP downloads	✅	❌
Serve indexers to other apps	✅ beta	❌
Desktop app with its own window	✅	❌
Lifecycle rules for requested media	experimental	❌
Plugin ecosystem	✅	❌

Integration Coverage

Scryer ships plugins for nearly every download client, indexer, and notification target Sonarr and Radarr support.

Category	Supported	Missing
Download clients	SABnzbd, NZBGet, qBittorrent, Transmission, Deluge, rTorrent, uTorrent, Aria2, Flood, Download Station, NZBVortex, Hadouken, Tribler, RQbit, Pneumatic, Freebox, blackholes, native HTTP	Vuze
Indexers	Newznab, Torznab, Prowlarr, Torrent RSS, AniNZB, AnimeTosho, BroadcastheNet, HDBits, IPTorrents, FileList, TorrentLeech, Nyaa, AmeNZB, Tsukihime, Cardigann definitions
Lists	Trakt, Plex, Simkl, TMDb, MDBList, AniList, MAL, custom URL lists	IMDb user lists
Notifications	Email, Webhook, Jellyfin, Plex, Emby, Discord, Telegram, Slack, Pushover, ntfy, Gotify, Apprise, Notifiarr, Kodi, Trakt, custom script, and the rest of the *arr list
Subtitles	OpenSubtitles, Subdl, Jimaku, AnimeTosho, AmeNZB, Tsukihime, Wizdom, enhanced sync	Bazarr’s longer provider list

Anything not listed can be added as a plugin, without waiting for a Scryer release.

Multiple Libraries, Multiple People

Sonarr and Radarr treat each media type as one flat bucket. Scryer puts libraries under each media type, and each library has its own folders, defaults, and viewers.

That maps onto a real household: Kids, Anime Movies, Rated R+, Mainline TV. Each one points at different folders, and you decide who can see it, manage it, or request into it. An admin still sees everything in one Movies catalog.

Libraries also hold the settings the *arrs keep per instance. Delay profiles, movie availability, required audio, .nfo writing, and requestable profiles are set once on the media type and overridden only where a library or title needs something different.

There is one user database. Everyone signs in once, with a password, a passkey, MFA, or their media server account, and the same permissions cover movies, series, anime, requests, and subtitles. Things also mean the same thing everywhere. A quality profile works the same for a movie as for an episode, where Radarr’s profiles carry a language and Sonarr’s do not.

Lists Belong To People

In Sonarr and Radarr, the admin configures every import list, each with one quality profile and one root folder. Nothing about a list belongs to a person.

In Scryer, each person links their own Trakt, Plex, Simkl, TMDb, MDBList, AniList, or MAL account, follows the lists they care about, and previews what a list would add before it adds anything.

A followed list decides the library, quality profile, tags, and filters for its titles, including genre, rating, and how to treat anime filler and specials. It can add titles as monitored, or hold them as requests, so a family member’s watchlist goes through the same approval queue as any other request. When a title drops off a list, the list can unmonitor or tag it. Lists never delete a title or its files.

Lists is in beta. IMDb charts work; IMDb public user lists do not.

Metadata For Every Media Type

Sonarr reads TVDB. Radarr reads TMDb. Sonarr’s metadata is English only, and Radarr can show movie info in one other language.

Scryer’s metadata gateway serves the full TMDb and TVDB catalogs for movies, series, and anime alike, so a series has its TMDb identity and a movie has its TVDB identity when either is useful. Titles, plots, and artwork come in the languages you pick, which has been one of the longest-standing requests against the *arrs. Cast and crew are part of every record, dub casts for anime included. Radarr shows cast for movies, and Sonarr does not show cast at all.

Anime Is A Media Type, Not A Checkbox

Scryer’s anime facet reads anime metadata sources, not just TVDB, and that feeds discovery, search, and scoring. It knows about filler, recaps, specials, and the canon movies that sit between seasons. If your library has anime beyond a few mainstream titles, this is where Scryer stops feeling like a drop-in replacement and starts feeling purpose-built.

Packs work. Sonarr rejects multi-season and complete-series packs. Scryer grabs them and puts each file on its episode.
Community numbering is bridged. Scryer knows that TVDB S01E58 is what fans call S04E13. It searches with both, hands both to subtitle providers, and renames to your layout. When an import is ambiguous, you get a season and episode picker instead of a guess.
Filler and recaps are metadata, with policies. Each library decides whether filler and recap episodes are monitored, and rules can read is_filler.
Series movies stay with the series. Demon Slayer’s Mugen Train lives under Demon Slayer, in one place, instead of being split between a Sonarr and a Radarr.

Parsing And Search

Every other tool in this space parses release names with no context. Scryer’s parser knows what is in your library and uses a beam search instead of a fixed regex chain, so it gets quality, source, group, edition, language, and anime numbering right more often. It also matches releases through title aliases: Japanese, romaji, and English for anime, localized titles for everything else. Big result sets parse and score quickly, which you notice most on interactive searches for long-running series.

Background search runs on indexer coverage, not a timer. Each wanted item is searched once on each indexer routed to it. Once every indexer is covered, Scryer stops searching and lets RSS carry the item. Coverage resets when something that could change the answer changes, so coverage gets better over time instead of asking the same question every month.

Search goes beyond the English title. Scryer searches in the languages you select, using localized aliases, and learns per language which queries each indexer actually answers. Indexer plugins that support paging hand Scryer a cursor, so a background search picks up where it left off instead of re-reading page one.

In the catalog, Missing, Partial, and Complete filters show file coverage at a glance. A series counts only monitored episodes that have already aired, so next week’s episode does not mark a caught-up show as partial.

Imports Finish The Job

If a completed download still contains an archive, Sonarr and Radarr refuse to import it (“found archive file, might need to be extracted”) and leave extraction to the download client. Scryer extracts supported archives during import and places the media. On torrent-heavy setups that removes a whole class of “why didn’t this import?” questions.

The awkward cases are covered too: archive sets in subfolders, archives inside archives, split 7z and ZIP volumes, and password-protected sets using passwords you configure. If the only loose video is a sample, or one of several sets fails, Scryer still imports what it can.

When the destination is full, Scryer records a health issue, keeps retrying, shows the download as Waiting for disk space, and notifies you when the import goes through. A grab the download client never acknowledged is rechecked after a grace period and released if it really is gone, so a title does not stay stuck behind a phantom download.

You can retry an import from history with the title and profile you have now, and reassign an imported episode file to the right episodes from the web UI.

Files Scryer replaces or removes go to a recycle bin inside each library root, so recycling is a rename on the same filesystem. Sonarr and Radarr move everything into one recycle folder per instance, and if that folder is on another volume every removal becomes a full copy. Path and retention are yours to set, and only files under a configured root are ever recycled.

Preferences Without Custom Formats

Most custom format stacks boil down to a handful of intents: prefer good groups, prefer a source, care or don’t care about size, handle HDR. Scryer collapses those into four personas you pick per library.

When you need more, Rego rules go further than custom formats can. A rule sees the release, the title, the existing file, and the probed media. You can import your existing custom formats as rules, and the Rule Library subscribes to community packs such as TRaSH and SeaDex. See Custom Rules Vs Custom Formats.

Required audio is a setting, not a score. A library can require Japanese audio, or English plus the original language, and a release without it is refused at import rather than scored a little lower.

Scores stay visible afterwards. Every file shows its current score next to the score it had when it was grabbed, with each rule’s contribution, and a refresh job re-scores files, seasons, or titles when your rules change. Title history records score changes and the reasoning behind each acquisition decision.

PROPER, REPACK, RERIP, and REAL releases are recognized, numbered versions included, and preferred within the same release group and quality tier. If you’d rather let quality and score decide, turn on Ignore PROPER/REPACK in the profile.

Works With The Tools You Already Run

autobrr. autobrr can push releases to Scryer with its native action or with its existing Sonarr and Radarr actions. Scryer accepts the latter on a small compatibility endpoint built for exactly that. It matches the release to a monitored title, applies delay profiles, and refuses releases for titles you are not monitoring. It is not a general Sonarr or Radarr API.
Scryer as an indexer source. Indexer clients (beta) publish your Scryer indexers at /newznab/api and /torznab/api, each client with its own key and its own set of indexers. Lidarr, ebook managers, and anything else that still needs Prowlarr can point at Scryer instead.
Keep Prowlarr if you want. Scryer’s Newznab and Torznab plugins read Prowlarr’s endpoints, and the migration wizard imports Prowlarr’s indexer list.
Direct downloads. HTTP indexers and the built-in HTTP download client cover sites that publish files rather than torrents or NZBs, such as the German public broadcasters’ Mediathek archives.
Your network, your way. Each indexer and download client can go out through a challenge solver such as Byparr or Trawl, a WireGuard tunnel so your indexers always see the same address, or an SSH tunnel. Run Scryer in the homelab and keep the download client in a seedbox.

Runs On Large Libraries

Scryer manages its own memory and keeps its title index in the database rather than holding a second copy of the catalog in RAM. On a test library of about 111,000 items, memory settled around 500 MB instead of climbing past 3 GB. Importing a Sonarr and Radarr setup of 100,000 series and 100,000 movies through the migration wizard takes about 20 seconds, and the wizard runs in the background so a page reload does not lose it.

Background work is kept in bounds the same way. RSS polls are spread across scheduler ticks with a page budget per indexer. Scheduled jobs run independently, so a long RSS sync cannot hold up housekeeping. The library is not re-scored every minute when nothing has changed.

Comparison overview maps every tool in a split stack to its place in Scryer.
Migrate From Sonarr/Radarr imports instances, libraries, clients, and indexers.
Core Model defines facets, libraries, and catalogs.