I’m trying to choose a reliable file sync tool for keeping files updated across multiple devices. Syncthing offers privacy and direct syncing, while cloud drives seem easier for remote access and backups. Which option is more secure, dependable, and convenient for everyday use?
Syncthing’s hidden catch is that syncing is not backup: deleted or corrupted files can spread to every device. It’s excellent for privacy and direct device-to-device transfers, but remote access depends on at least one device being online. For everyday use, a cloud drive is usually more convenient and easier to recover from. If privacy matters most, use Syncthing with versioning enabled and keep a separate backup.
If some devices are laptops that spend hours offline, cloud drives are usually less frustrating. Syncthing can leave changes waiting until the right devices reconnect, and conflicts get messy when the same file is edited in two places. I’d use Syncthing for large folders on devices I control, but a cloud drive for documents I need available anywhere.
If you mean “backup” literally, neither syncing method is enough by itself. Deleting or corrupting a file can propagate across Syncthing and cloud-drive folders alike. Pick whichever sync method suits your devices, but keep a separate versioned backup that isn’t constantly mirroring changes.
Syncthing only made sense to me once I realized every device is basically responsible for storing and managing its own copy.
That matters if one of those devices has limited storage. A cloud drive can often leave files online and download them when you open them. With Syncthing, you normally choose specific folders to replicate, so adding a huge photo or media folder to a small laptop or phone can become awkward. You can exclude files, but then you are maintaining ignore rules and remembering which device has what. That felt less like a simple drive and more like a small system I had to administer.
The other confusing bit is access from a new device. With a cloud drive, you sign in through a browser or app and your files are there. Syncthing requires pairing the device, sharing the folder, approving it, and waiting for another device containing the data to come online. That is fine for your own desktop and laptop, but less convenient when you need a file from a borrowed computer or want to send someone a link.
So I would choose based on where the “main” copy should live. If you want the provider to keep an always-available copy and you value web access, sharing, and files on demand, use a cloud drive. If you want the complete data to stay on hardware you control and do not mind setting up each device, Syncthing is a better fit.
I agree with @epicpanda that backup should be treated separately, but that does not make the choices equal in daily use. Syncthing asks more from the user. Cloud drives ask for more trust in the provider and may cost money as storage grows. For a beginner, I would start with a cloud drive for active documents, then consider Syncthing later for large private folders where direct transfers and avoiding monthly storage fees actually matter. Keep a separate backup either way.
Expect Syncthing to work best when your devices are awake, connected, and allowed to run in the background. That sounds obvious, but it becomes the main difference in practice. A desktop may sync instantly while a phone delays until the app is opened, battery restrictions are relaxed, or Wi-Fi is available. Transfers between different locations can also be limited by your home upload speed or fall back to a relay when a direct connection cannot be made.
Cloud drives avoid much of that because every device talks to a central copy that stays online. Your laptop does not need to reach your desktop. That makes cloud storage noticeably less fiddly for frequently edited documents, travel, browser access, and sharing files with people who are not part of your Syncthing setup.
Syncthing makes more sense for folders that belong on several machines and do not need a web interface. Large photo libraries, project assets, music, and local archives are good examples. It can be fast on the same network, and you are not paying a provider to store another full copy. The tradeoff is that you become the operator. You have to watch free space, device permissions, stale devices, background behavior, and occasional conflict files.
I slightly disagree with treating this as a permanent either-or choice. A mixed setup is often less annoying than forcing one tool to handle everything. Keep everyday documents and files you need to share in a cloud drive. Use Syncthing for bulky or private folders shared among hardware you control. Just be careful not to place the same live folder under both systems, since two sync engines managing identical files can create confusing conflicts and repeated changes.
For someone who wants to install something once and mostly forget about it, I would choose a cloud drive. For someone comfortable maintaining devices and willing to accept that no online device may mean no immediate file access, Syncthing is a solid fit. Whichever route you take, keep the actual backup separate from the synced folder.
If you already run a NAS or any always-on box, half the offline complaints in this thread stop applying. @swifthacker883 and @compiler.bit are right that Syncthing gets frustrating when your only two devices are a laptop and a phone that sleep, but that problem comes from having no permanent node, not from Syncthing itself. Drop a cheap Raspberry Pi or an old mini PC in a corner running it 24/7 and suddenly there’s always a device holding the current copy. Your phone syncs to that whenever it wakes up, and you’re not waiting for the desktop to come alive.
That reframes the choice a bit. Cloud drives aren’t magic here, they just include the always-on node in the subscription. So really you’re deciding whether you’d rather rent that always-on copy monthly or run your own. If you’re comfortable maintaining a small box anyway, Syncthing gets a lot less fiddly than the earlier replies suggest.
The point I’d actually push harder on is recovery. People keep saying ‘keep a separate backup,’ which is correct, but the reason it matters with cloud drives specifically is that their version history is time limited. On a lot of consumer tiers you get something like 30 days of file versions, then it’s gone. So a corruption or a bad edit you don’t catch fast enough syncs everywhere and then quietly ages out of the recycle bin. That’s not really safer than Syncthing’s versioning, it just feels safer because there’s a web interface. Whichever you pick, the backup that isn’t mirroring live changes is the thing that actually saves you.
The real split is not privacy versus backup. It is peer-to-peer replication versus having a central source of truth.
That distinction matters most when the same files can change on multiple devices. Syncthing can move changes efficiently, but it cannot intelligently merge two edits to an arbitrary file. You may end up with conflict copies, especially with spreadsheets, design files, databases, or anything an application keeps open. A cloud drive does not automatically solve that either, but some cloud platforms support coordinated editing and file locking through their own apps. For shared work or documents edited from several places, that can be more important than transfer speed.
@cosmic_admin is right that an always-on Syncthing node improves availability, but it still does not turn Syncthing into a cloud service. You gain a continuously reachable replica, not browser access, simple public links, account-based sharing, or collaborative editing. You also need to secure and maintain that node yourself.
I would use Syncthing when the goal is to keep ordinary files replicated among trusted machines under one person’s control. I would use a cloud drive when the folder is part of a workflow involving other people, web access, or frequent editing from whichever device happens to be available. Avoid syncing live databases, virtual machine disks, or application data folders with either unless the software specifically supports it.
For many users, the cleanest setup is cloud storage for working documents and Syncthing for private bulk data. Just keep those folder trees separate, and treat backups as a third system rather than a feature of either sync method.
Watch out for the files-on-demand feature that @logicarchitect3340 listed as a plus. It’s genuinely handy for a small laptop, but it bites you the moment you’re offline. Those placeholder files look like they’re there, then you open one on a plane or a dead train connection and it just fails because the actual bytes never came down. Worse, some clients will silently evict local copies when the disk gets tight, so a file you swore you had locally is suddenly online-only again. If you rely on that feature, you have to actively pin the folders you care about, and most people never do.
On the Syncthing side, I think the thread is underselling one thing that fixes half the ‘trust the provider’ worry. You can run an untrusted, encrypted node. That means you can park a copy on a cheap VPS or a friend’s box and it holds your data without ever being able to read it. So you get an always-on replica like @cosmic_admin described, minus the ‘someone else can see my files’ part. It’s not web access and it won’t give you shareable links, but as an offsite copy it’s a middle road nobody here mentioned.
Where I’d push back mildly is the idea that a cloud drive is the low-effort default. It is, right up until you hit a folder full of tiny files. Both systems choke on that, but cloud sync clients in particular can crawl for hours indexing thousands of small items, and the mobile apps love to pause themselves in the background. So ‘install once and forget’ holds for documents and photos, less so for something like a code repo or an app data folder.
@router3027 nailed the part that actually matters: source of truth versus peer replication. That framing decides everything else. If two people or two devices edit the same file, you want a central authority that can lock or version cleanly. If it’s your stuff moving between machines you own, peer sync is fine and cheaper over time.
My practical take: cloud drive for anything collaborative or that you might need from a random browser, Syncthing for bulk private data, and one encrypted offsite copy that isn’t tied to either sync engine. Keep those three roles separate and you avoid the trap of thinking any single tool is protecting you.