A modern file downloader with parallel chunking, resumable transfers, and cross-platform sync. CLI and web UI — one unified database.
Everything you need from a download manager, without the bloat.
Splits files into chunks and downloads them concurrently. Up to 16 parallel streams for maximum throughput.
Never start from scratch. Pause and resume downloads across sessions with crash recovery.
Intercept YouTube links and choose quality — 4K, 1080p, 720p, or audio-only. Powered by yt-dlp.
Use the CLI or web UI. All share the same SQLite database — instantly synced.
Capture downloads from Chrome, Edge, or Firefox. Right-click any link and send it straight to Grabr.
Real-time download speed with EMA smoothing, ETA calculation, and a chunk-level progress visualizer.
SQLite-powered. Every platform reads and writes to the same ~/.grabr/grabr.db file. No cloud, no tracking.
Configure max concurrent downloads. Jobs automatically queue and process as slots become available.
Choose your interface. The same downloads, everywhere.

Full-screen terminal dashboard

Web-based UI (Hono + Bun)
# Install globally
bun install -g @linuxctrl/grabr
# Verify installation
grabr --version
# → 1.0.11Route browser downloads directly to the Grabr daemon.
Coming soon
Install manually from sourceComing soon
Install manually from source