A bookmark points at a picture. A file is the picture.

That sentence is the entire difference and everything below is consequences of it. Raindrop is a bookmark manager — an extremely good one, free forever at the tier most people need — and its unit is a URL. Coii Ref's unit is a file that is already on your disk, indexed where it sits. If your references live as links you have collected from the web, Raindrop is the right tool and this page should end with you keeping it. If they live as tens of thousands of actual files across four folders and two drives, a bookmark manager was never going to be the answer and the upload allowance is where you find that out.

What Raindrop is, with today's numbers

Read off their own pages on 3 September 2026.

Raindrop is free to use indefinitely, with unlimited bookmarks, collections and devices on the free plan — which is a more generous free tier than almost anything else in this comparison series, and worth saying before anything else. It runs as browser extensions for Chrome, Firefox, Safari and Edge, plus apps for Windows, Mac, iPad, iPhone and Android.

Their premium features page is where the numbers that matter for images live:

• File uploads: 100 MB a month on free, 10 GB a month on Pro.
• Pro adds full-text search across the content of a bookmark, including PDFs, EPUB files and YouTube transcripts.
• Pro adds a permanent web archive — saved copies of the pages themselves, so a dead link is still readable.
• Pro adds a broken link checker, a duplicate finder, reminders, annotations, scheduled exports of the whole library, and an assistant that answers questions about what you have saved.

For price, the Mac App Store listing for their Mac app — published by Rustem Mussabekov, requiring macOS 12 or later — shows PRO as in-app purchases at $2.99 monthly and $27.99 yearly. Their help pages confirm both billing options exist and put the yearly discount at roughly twenty percent.

Coii Ref: $19 USD once, three Macs, updates included, a thirty-day trial with every feature and no card and no account, macOS 11 Big Sur or later, native on Apple Silicon and Intel.

The question the keyword is actually asking

People searching "raindrop for images" are usually one of two people.

The first has been using Raindrop for a while, likes it, and is wondering whether the reference pile can move in too. The honest answer is the upload allowance: 100 MB a month free, 10 GB a month on Pro. A single afternoon's shoot can exceed the free figure, and a four-year reference archive is normally somewhere between fifty gigabytes and several hundred. Raindrop is not attempting to be that, and the numbers say so plainly rather than by implication.

The second has thousands of images already on a disk and is looking for anything at all that will make them findable. For that person the upload model is not a limit to work around, it is the wrong direction entirely: the files do not need to go somewhere, they need to be readable where they are.

There is a third case worth naming because it is the one Raindrop wins outright. If what you actually collect is articles, tutorials, colour theory posts, portfolio sites and the occasional PDF, then the reference is the page and not an image on it — and a tool that keeps a permanent copy of that page, and searches the words inside it, is doing something no local image index can do at all.

Where Raindrop is clearly the better tool

It saves from the browser. One click, four browsers, both phones. Coii Ref has no extension and no clipper; it reads folders and has no concept of a web page. Every guide on this site about collecting from the internet is about what happens after the file has landed on disk, which is a real gap, not a stance.

It holds everything, not just pictures. Links, articles, PDFs, EPUB, videos, notes on all of it. Coii Ref indexes images and video and nothing else — no PDFs, no documents, no web pages. For a practice that is half reading, that alone decides it.

Full-text search inside the thing. Searching the words in a saved PDF, or a YouTube transcript, is a category of search we cannot offer, because we search what is true of a file rather than what is inside it.

The archive. A permanent copy of a page means the reference outlives the site. That is a genuinely valuable property and it is the reason to pay for Pro even if you never upload anything.

Everywhere, on everything. Windows, Mac, both mobile platforms, the web. Coii Ref exists on the Mac that holds the folder and nowhere else.

And the free tier is real. Unlimited bookmarks and collections, no expiry, no nagging count. Recommending against it would be silly.

Where the local side takes over

The case for a file index is narrow and it is about a specific pile of material: the images you already own, that nobody tagged, that are not going into anybody's service.

Coii Ref points at a tree and reads it. Files are never copied, moved or renamed. Everything it writes — the index, the thumbnails, the video proxies — goes inside the folder you pointed at, which means the backup you already run covers the library, and moving the folder moves the workspace:

Then the search. One line, AND across keys, OR inside a comma list, no parentheses, and anything the parser does not recognise falls through to free text rather than throwing an error:

The keys are type ext tag rating is has no in width height added modified. The reason this matters more than it sounds is that none of it required tagging. Size, format, folder, date added, date modified and rating are facts the disk already holds, so on the first day of an index — before you have typed a single tag — a four-year archive is already interrogable.

Saved views keep the text of a query and re-parse it when opened, rather than storing a list of members. Rename a tag and every view that mentioned it follows.

The table

Seven rows in bold on their side, two of which — the free tier and the cross-platform reach — are the ones that decide it for a lot of people. This is a narrow overlap dressed up as a comparison, and the useful reading of the table is which column has the row you actually care about.

What an index pass actually does

Worth spelling out, because "it indexes your folders" is the kind of sentence that could mean almost anything.

You point at a parent folder. It walks the tree and reads what each file already says about itself: kind, extension, dimensions, EXIF, colour, when it arrived, when it last changed. It writes thumbnails so a grid of ten thousand things scrolls, and proxies for video at canvas resolution so a board of clips behaves like a board of stills. It writes all of that inside the folder you pointed at. It does not copy a single original, rename one, or move one.

Then it stops walking and starts listening. Filesystem events become the smallest re-scan that makes the database true again, so a rename in Finder stays the same item with the same rating and notes attached rather than becoming a new one and losing them, and a folder of new work from a shoot appears without you asking for anything. A drive that spent a month unplugged reconciles when it comes back instead of starting from nothing.

That watcher keeps running after the trial ends, which is deliberate: an index that quietly rots because you did not buy is worse than no index.

Video, and the row nobody checks

A clip is second-class in most tools: different folder, different application, absent from the search that would have found the thing it shows. Raindrop will happily store a video file, within the upload allowance, and that is a reasonable thing for a bookmark manager to do.

On the local side video goes through the same pass as everything else. Same tags, same ratings, same query line — type:video ext:mov rating:>=4 is an ordinary sentence rather than a mode — and the proxies mean a canvas of footage scrolls rather than stutters. If a good part of what you study moves, that is one library instead of two.

A small note on duplicates, since Raindrop's Pro tier has a finder for them and people ask. Coii Ref has no duplicate detector. What it has instead is a search that makes duplicates visible rather than resolving them: the same file in two folders shows up twice in a grid, and ext:jpg width:>=3000 over a tree is usually enough to see that you saved the same wallpaper four times in 2023. If de-duplicating a large pile is the job in front of you today, that is a different tool than either of these.

Boards, and the format argument

The second half of Coii Ref is the canvas, and it is where the difference stops being about scope and starts being about what you own.

A board saves as a .coii-canvas file in a folder you chose: pretty-printed JSON naming the files it refers to and where each sits. It opens in a text editor. It diffs in git. It syncs with whatever you already sync with. There is no export command because there was no import, and there is no account holding the only copy.

If you like that reasoning you will find it again, made by other people, in Obsidian's canvas format — which is genuinely open, published as a spec, and worth reading about whether or not you end up here. The idea that a document should outlive the application that made it is not ours; it is just unusually easy to check in this case.

What Coii Ref does not have

Plainly, so nothing in the trial is a surprise. No accounts, no cloud, no sync, no sharing links, no collaboration and no mobile app. No browser extension and no clipper. No web archive, no reading list, no highlights on articles. No PDFs, EPUB, documents or 3D formats — images and video only. No full-text search of anything, because there is no text to search. No automatic tagging and no search by meaning: the query line is literal. No search by colour — colour is read into the index, but there is no colour filter. And macOS only, version 11 or later, which rules us out entirely for anyone working across two kinds of computer.

The network list is two items long: a licence check when you press Activate, and one request at launch read only for its Date header, so that winding the clock back cannot mint free trial days. No analytics, no crash reporting, no update check.

Running both, which is the recommendation

The seam is clean here, cleaner than in most of these comparisons, because the two tools barely touch.

Raindrop holds the links. Articles, tutorials, portfolio sites, the PDF somebody sent, the YouTube breakdown you will rewatch. Anything whose value is the page rather than an image on it. That is what a bookmark manager is for and there is no reason to move it.

A folder on your Mac holds the images. When a reference stops being something you might read and becomes something you will draw or shoot or grade from, it becomes a file in a folder whose name will still mean something in three years. Then one index pass over the parent, and everything in there is searchable in the same breath — including the four thousand files from before you had any system at all.

Two habits make that work. Save the source URL into the notes field on a file when it matters, so attribution survives the download. And put the good stuff somewhere better than Downloads on the day it arrives, which is the only manual step in the whole arrangement and takes about four seconds.

Trial, and what stops

Coii Ref's trial is thirty days with every feature switched on, no card, no account. On day thirty-one, if you have not bought, one category of action stops: the one that grows the library — registering a new folder, running the importer, starting an index pass by hand. Everything that reads keeps working permanently, including search, tags, ratings, saved views, canvases and thumbnails, and the folder watcher carries on, so an unpaid copy does not let the index go stale behind your back. It just stops taking on more.

Raindrop's equivalent is simply its free tier, which does not expire at all — a different and arguably kinder shape, and one it can afford because the expensive part of what it stores is optional.

Who should pick which

Raindrop if what you collect is the web: links, articles, PDFs, pages worth archiving, things you want searchable by their words, on every device you touch, for nothing or for $27.99 a year. Nothing on this site competes with that and nothing here is trying to.

Coii Ref if what you collect is already files: tens of thousands of images and clips on your own disk, unfiled, unfindable, and far past any upload allowance. $19 once, indexed in place, searched from one line, arranged on canvases that are files.

Both, in the division described above, which is what most people reading this already half do.

Neither, if the pile is small. Finder and a sensible folder tree genuinely work up to a couple of thousand images, and the moment to buy anything is the first time you cannot find a picture you are sure you saved.

If it is the image side you are weighing and you want to see the field: Eagle is the well-known desktop option and takes your files into a managed library rather than reading them in place; Savee and Are.na are the two subscription services people most often use as visual bookmark managers; and Finder tags is the free thing already on your Mac, which is worth understanding before you replace it.

Otherwise: the library has the rest of the comparisons, and the trial is thirty days with every feature, no card and no account.