Pluton

Capabilities

• On-device encryption and compression Files are encrypted and compressed on your machine before upload using the established Restic format, so storage providers hold ciphertext they cannot read.[1]

• Incremental scheduled backups Plans run on a scheduler and transfer only changed blocks, which keeps storage use and transfer times low after the first full snapshot.[1]

• 3-2-1 replication across backends A replication feature keeps copies of backup data on multiple storages, making the classic three-copies, two-media, one-offsite pattern practical without manual copying.[1]

• More than 70 storage integrations Local disks, SMB, SFTP, FTP, and WebDAV sit alongside Backblaze B2, Amazon S3, Google Drive, OneDrive, Dropbox, Azure Blob, Google Cloud Storage, Storj, and others, so existing storage can be reused.[1]

• Smart retention policies Flexible keep-and-prune rules rotate old snapshots automatically, which bounds storage growth without hand-deleting archives.[1]

• Guided restore with snapshot browsing Snapshots can be explored in the interface like a file manager, and restores target the original location or a custom path; whole snapshots download as TAR files for offline access.[1]

• Lightweight self-hosting The program runs from a Docker image or a standalone executable, idles near 50MB of memory, and suits low-resource hardware including single-board computers.[1][2]

• Operational safeguards Real-time progress and detailed per-job logs aid diagnosis, automatic retries handle transient failures, two-factor login protects the console, and notification integrations report job outcomes.[1]

Who it fits

• People consolidating scattered cloud storage will get the most out of it: files already sitting across Google Drive, S3, and a NAS become backup destinations under one encrypted plan instead of requiring another paid vault.[1][2]

• Privacy-minded households benefit because on-device encryption means the storage provider never sees file contents, which matters for personal documents kept on third-party clouds.[1]

• Tinkerers with a spare machine or container host can put existing hardware to work, since Docker and standalone builds carry a small idle footprint.[1][2]

Boundaries to check

• Teams needing proven decade-scale restores should wait: backup software is judged by recoveries years later, and a project started in late 2025 cannot yet show that record.[1][2]

• People who want fully managed backups should look elsewhere: storage accounts, updates, monitoring, and restore testing remain your responsibility, with no vendor-hosted plan to fall back on.[1][2]

• Environments requiring SSO or compliance paperwork will find a gap: only two-factor login is documented, with no stated SSO, audit reporting, or enterprise support plan.[1]

Fit and compatibility