Built to run on your own machine
FluxSave is an original, open-source implementation with no paid APIs, no subscriptions and no external services in the critical path.
Modular extractors
Each platform is an isolated module implementing canHandle, analyze, getFormats and download. Unsupported links say so instead of pretending.
Smart format engine
The FormatResolver maps your requested quality to the best available source streams and prefers remux over transcode whenever the containers are compatible.
Job queue with real states
Queued, analyzing, downloading, processing, ready, failed, expired, cancelled — with live speed, ETA and per-job cancel, retry and remove.
Audio, video, images, GIFs
MP4 and WebM video, MP3/M4A/WAV/FLAC audio, JPG/PNG/WebP images and GIF — only ever offered when the source actually provides them.
Subtitles and chapters
Download publicly available SRT or VTT tracks, or embed them into the output file. Nothing is invented or auto-generated in the cloud.
Temporary by default
Every produced file carries created_at and expires_at. A cleanup worker removes expired files on a 30 minute TTL you control.
Private mode
History off by default, no accounts required, no third-party analytics, no telemetry. Optional local sign-in if you share your instance.
Zero-cost stack
SQLite, local filesystem, an in-process queue, FFmpeg and yt-dlp. Redis, Postgres or S3 can be added later without a rewrite.
Extractor registry
Supported sources
| Platform | Media | Status |
|---|---|---|
| YouTube | video, audio, subtitles | available |
| Vimeo | video, audio | available |
| SoundCloud | audio | available |
| X / Twitter | video, image | available |
| video, image, gif | available | |
| TikTok | video, audio | available |
| video, image | available | |
| Twitch VOD | video | available |
| Direct file | video, audio | available |
Public content only. FluxSave does not bypass DRM, logins, paywalls or any other access control, and you are responsible for downloading only what you own or are authorized to download.