Is an AI Website Builder Really Free? What Free Buys
Free gets you in the door, not out of it. Zugo's free plan carries 5 credits as a one-time grant and a fresh build costs 6, so the free tier is a look at the product rather than a launch. Publishing a real site starts on a paid plan, and the connected services bill you separately.
That is an uncomfortable sentence to open with, and it is the accurate one. This post lays out where every cost sits, including the ones that are not on our pricing page because they are not ours to charge.
What does "free" usually mean in an AI builder?
Three different things, and the word does not distinguish them. It can mean a trial with a fixed allowance. It can mean a permanently free tier with capabilities removed. Or it can mean free to build and paid to publish, which is the one that catches people out after the work is done.
The version that matters to you is whichever one applies at the moment you have something you want to show somebody. A tool that is free right up until the point of usefulness has not saved you anything, it has just moved the decision later, after you are invested.
So the test to apply to any builder, ours included, is simple: what does it cost to get a working thing at a public address that a stranger can open? Everything before that point is rehearsal.
What does Zugo's free plan actually include?
Five credits, granted once. Because a fresh build costs 6, those credits do not complete a full generation on their own, and we would rather say that here than let you discover it mid-sentence.
What the free tier is genuinely good for is looking at the product before you decide. You can see the interface, read the templates, and understand how describing a project translates into a build, which is more information than any pricing page carries.
There are 25 templates in the product, 5 of them games. Reading through them is the fastest way to calibrate what "describe it and it builds" actually means in practice, and it costs nothing at all.
What do the paid plans cost?
Two plans, and prices are per action so the arithmetic is available before you commit.
| Action | Credits |
|---|---|
| Single build: site, app, or game | 6 |
| Edit to an existing build | 3 |
| Multi-page platform, first three pages | 12 |
| Each page after the first three | 3 |
| Hi-Fi build | 12 |
| Hi-Fi edit | 6 |
Pro is $25 per month with 200 credits. In actions that is roughly 16 full platforms, or 33 quick builds, or 66 edits if editing is all you spend on. Business is $99 per month.
Most real projects are one build followed by a run of edits, so the mixed number is the one to plan against. A landing page built once and refined eight times is 6 plus 24, which is 30 credits, and a Pro month holds several projects of that shape at the same time.
What costs money that is not credits?
The connected services, because they are your accounts rather than ours. This is the part that is genuinely invisible until you need it, so here it is in one table.
| What you connect | What it does | Who bills you |
|---|---|---|
| Supabase | Database, sign-in, file storage | Supabase, on their plan |
| Stripe | Payments and subscriptions | Stripe, per transaction |
| Resend | Transactional email | Resend, on their plan |
| Vercel | Deployment to your own account | Vercel, on their plan |
| GitHub | Export of the source | GitHub, on their plan |
| Google Analytics | Traffic measurement | |
| Your custom domain | The address people type | Your registrar, yearly |
Several of those have free tiers of their own that comfortably cover a small project, so the practical bill for a first site is often the domain and nothing else. The point of the table is not that these are hidden fees, it is that they are real accounts under your name, and that is the same reason you keep them if you ever leave.
A published project lives at <slug>.zugo.run and a custom domain connects on top of that. If you never buy a domain, you still have a working public address.
Is the free tier enough to evaluate the product?
For understanding the workflow, yes. For judging output quality on your own idea, not really, and it would be dishonest to imply otherwise when 5 credits do not complete one build.
The realistic evaluation is one paid month spent deliberately. Build the project you actually care about, edit it four or five times until it stops embarrassing you, publish it, and put the address in front of two people who are not you. That sequence costs a fraction of a Pro month and answers the only question that matters, which is whether the output is good enough for your purpose.
If the answer is no, you have spent $25 and a weekend to find out, which is cheaper than most alternatives to finding out. If the answer is yes, you already have the thing published.
How does this compare with the alternatives?
Honestly, and the comparison depends on what you were going to do instead.
| Instead of a builder | What it costs | What you get that a builder does not |
|---|---|---|
| A site template you buy once | A one-off fee, then your time assembling it | Full manual control over every element |
| A freelancer | Considerably more per project | Judgement about your business, not just execution |
| An agency | More again, on a schedule | A team, a process, and accountability |
| Building it yourself in code | Your hours | Exactly what you wanted, eventually |
None of those are bad choices, and two of them are better choices for a complex product. Zugo does not replace a development team when the requirement is genuinely intricate, and a person who understands your market is worth more than any generator when the problem is deciding what to build rather than building it.
Where a builder wins is the first version, and the second, and the seven experiments you would never have paid a freelancer to run. The detailed arithmetic against those alternatives is in how much an AI website costs.
What are the honest limits of the free tier?
Three, stated plainly.
Five credits do not finish a build. It is a look, not a launch. Anyone who tells you a free tier of any AI builder produces a finished business asset is selling something.
Verification is a check, not a warranty. Every build is booted in a sandbox before it reaches you, and a build that fails to load is reported as a failure rather than handed over. That lowers the risk of receiving a blank page. It does not remove it, and it does not judge whether the content is any good.
Specific business logic costs edits. Unusual rules arrive through a sequence of edits rather than one perfect prompt, and edits are 3 credits each. Budget for the passes rather than assuming the first build is the last one.
The straightest way to settle whether the paid plan is worth it is to price your own project against the table above before you sign up. The full per-action breakdown lives in AI app builder pricing, and what happens when an allowance runs out is covered in what happens when I hit limits. When you have a number you are comfortable with, describe the project at Zugo and spend the first credits on the thing you actually want.