The useful finding of this comparison is a null result, so it goes at the top rather than buried under nine hundred words of throat-clearing: from everything this site can check, CnFans and MuleBuy are nearly the same object, and we cannot tell you which is better.
That is not a cop-out. It is what the evidence looks like when nobody has run the measurement, and it is a more useful thing to know than a coin flip dressed up as a verdict.
They speak the same URL language
Both agents want the same two things to open a product: a numeric marketplace item id, and a platform token drawn from an identical three-word vocabulary — Weidian, Taobao, 1688. Both fall back to accepting an entire encoded source URL when the id cannot be parsed.
The differences we can actually see are cosmetic:
- CnFans puts the product on a product path with the platform parameter first; MuleBuy uses the same path with a trailing slash and the id first.
- The referral parameter is spelled the same way in both, with different numbers in it.
That is the whole of the visible difference in how a buy link is constructed. Paste any Weidian, Taobao or 1688 item into the decoder and the two links it produces for these agents will be near mirror images.
What that similarity does and does not mean
A third agent, JoyaGoo, uses the same grammar again. Three front ends sharing a parameter vocabulary suggests a shared template, a shared vendor, or one interface that got copied — and it suggests nothing more than that.
We have established no corporate relationship between any of them and you should not infer one. Warehouses, staff, forwarder contracts, photographers and dispute handling are where an agent actually differs, and none of that is visible in a URL. Two shops can run the same checkout software and be entirely different businesses to deal with.
The cost stack is identical in shape
Both build an order the same way every agent does:
- Item price plus domestic Chinese shipping to the warehouse.
- A service or purchase fee for buying on your behalf.
- Storage — usually free for an opening window, then charged per day.
- International freight, quoted only once the parcel is packed, on chargeable weight, meaning the greater of the real and volumetric figure.
- Add-ons: extra QC photos, box removal, repacking, insurance and declared-value options.
So when someone tells you one of these two is cheaper, ask which line they measured. A lower headline service fee with a worse freight rate is a worse deal on almost any real haul, because freight is usually the largest number on the invoice. The only comparison worth making is a matched basket: same items, same weight, same destination, same week, all-in.
We have not run that comparison, so we are not printing percentages for either. Their own fee pages are the binding source, and the indicative shipping bands in how to buy reps apply equally to both, since both resell similar forwarder capacity.
Neither has produced an incident here
This site runs a checker against the two URLs it actually emits per agent — the referral link and a real product buy link — because homepages mislead in both directions. Its design was shaped by two specific false alarms: a European 451 on one agent's homepage while its buy link served fine, and two other agents that answer 403 to any bare robot request while serving browsers normally.
Neither CnFans nor MuleBuy is among the hosts that forced those rules. That is a small, thin positive: it means our monitoring has not had to make excuses for them. It is not uptime data, it is not a service record, and it should carry very little weight in your decision.
What would break the tie
The measurements that would let this page pick one, none of which exist here:
- A matched basket priced at both in the same week, freight included.
- Order-to-warehouse timings across enough orders to survive one bad month.
- QC photo sets of the same item from both, compared side by side.
- Dated dispute outcomes — refunds granted, refused, ignored — with the threads behind them.
- Payment success rates by country, which is where agents quietly differ most.
Until that exists, both sit unranked, alongside the other six. That is the same rule this site applies to a seller carrying no votes: not knowing is not a grade, and the method page exists so that nobody can buy their way past it.
How to choose anyway
When two options are indistinguishable on the evidence, choose on something you can verify yourself in ten minutes:
- Payment. Open both, and check which methods each accepts from your country. This is the difference most likely to actually bind, and it is the easiest to check.
- The interface in your language. You will be reading warehouse notices and QC photos in it for weeks.
- Existing links. If your Discord or your saved finds are already full of one agent's links, converting them is friction you can simply not have.
Then start small. Buy two or three cheap things rather than one expensive one, inspect the QC photos properly while the parcel is still returnable, and treat the first order as the one where you learn the process.
Pick something real to try it on from the shelf, or from a seller with quoted evidence behind them on the seller pages. The other pairing people ask about is covered in KakoBuy vs Superbuy, where there is at least one documented incident to talk about.