ToolScore

rclone

rclone project · backups · reviewed by Stan K. · Source: data snapshot · data as of 2026-11-12

Not a backup engine — the transport: the pipe your scripts and half the tools on this board ride to B2/e2/S3. Fresh releases, zero drama.

#8 of 12 in Backups

Try rclone for $0
Real price
$0 · OSS
open-source
Free tier
Yes
$0 — MIT. v1.76.0 (2026-11-06), six days before this snapshot. Fresh releases, zero drama.
Limits
4.0/10
$0 (MIT) · v1.76.0 (2026-11-06) · fresh releases, zero drama
Verdict
THE TRANSPORT
Not a backup engine — the transport. Retention and versioning are whatever the tool above it does; the ransomw

ToolScore breakdown — why 70/100

Price / Value
10.0
Token & Rate Limits
4.0
Speed / Latency
5.0
Privacy & Retention
5.0
BYOK (Own API Key)
9.5
Output Quality
6.0
Stability
9.5
Ease of Setup
6.5
Billing Transparency
9.0
Support & Community
6.0

10 criteria × 10 points. Low pillars are visible on purpose — that's the point of a score you can audit. Methodology →

What it does well

  • ▸Fresh releases, zero drama — v1.76.0 landed 2026-11-06, six days before this snapshot
  • ▸The pipe half the tools on this board ride to B2/e2/S3 — including as a backend for Restic and Kopia
  • ▸MIT-licensed and everywhere

Where it costs you

  • ▸Not a backup engine — no retention, no versioning, no ransomware story of its own (“—”, transport only)
  • ▸Sync is not backup: it moves bytes with the fidelity of the command that invoked it — immutability must come from the engine above or Object Lock below

Hidden traps — the stuff the pricing page doesn't say

medium riskIt's the pipe, not the vault

rclone moves bytes; it does not retain them. The ransomware column is “— (transport only)” — a sync is not a backup, and immutability has to come from the engine above it (Restic/Borg append-only) or the bucket below it (Object Lock on B2/e2).

Try rclone

Know the limits before you pay: read the traps above, then decide with eyes open.

Visit official site

Affiliate link where available — never affects the score.

What the advertised price hides
$0 — MIT. v1.76.0 (2026-11-06), six days before this snapshot. Fresh releases, zero drama.
Sources & freshness

Data as of 2026-11-12. Prices and limits change monthly — report anything stale. How we verify: methodology.

Closest alternatives

rclone review

In-depth analysis · last updated 2026-11-12

The pipe

rclone appears on this board in exactly one role, and the hub names it: not a backup engine — the transport. It is the pipe your scripts and half the tools above ride to B2/e2/S3. Restic lists it as a backend; Kopia lists it as a backend; your cron jobs list it as the command that moves the bytes. MIT-licensed, v1.76.0 (2026-11-06).

Why it still scores 70

Because in its role it is exemplary: fresh releases, zero drama — v1.76.0 shipped six days before this snapshot. The score is not “bad backup engine”; it is “excellent pipe, judged against a pillar set where retention and immutability are the product”. On those pillars rclone deliberately has nothing: its ransomware-defense cell is “— (transport only)”.

Where it fits in a real stack

The honest architecture on this board uses rclone as plumbing: engines (Restic, Kopia, Borg via mounts) own retention and append-only or Object Lock; rclone owns movement between endpoints. Used that way it inherits the safety of whatever is above and below it. Used alone — sync as a “backup” — it faithfully replicates mistakes with the same speed it replicates data.

How the pillars land

Price/Value 10 — $0, MIT, and it multiplies the value of everything else on the board. Retention & versioning 4 — the pillar it simply does not have: the hub lists its ransomware defense as “— (transport only)”, and retention is whatever the tool above it does. Speed limits sits at the neutral 5 — no throughput numbers are verifiable in this snapshot; “the pipe” is the job description, not a measurement. Encryption model sits at the neutral 5 — no cipher facts in this snapshot. BYO storage (B2/S3) 9.5 — it is how half this board reaches B2/e2/S3 at all. Feature set 6 — scored as what it is: a transport with no engine features, and a backend dependency of the engines that do. Track record 9.5 — “fresh releases, zero drama” is the hub's own verdict, backed by a release six days old. Ease of use 6.5 — excellent for scripters, and only for scripters; there is no backup product here to be easy at. Pricing transparency 9 — MIT, dated releases, and honest about not being a backup engine. Support & docs 6 — docs and repo verified; community support only.

Verdict — 70/100

Keep it installed, keep it updated, and never let it be the whole plan. As the transport under an engine with append-only or over a bucket with Object Lock, rclone is unbeatable plumbing; as the backup itself, it is a sync — and a sync is not a backup.

From the insights