Blink.new Review: AI App Builder, Pricing & My Test Results

Blink.new, also written Blink, is an AI app builder. You describe an app in a chat. It writes the pages, sets up a hosted database and member log-in, and puts the app online on its own address. To change something, you send another message, and you see the result in a live preview.

You pay with credits, so you should watch the meter. The Free plan gives 5 a day, and the paid plans run from $25 to $200 a month. I used the Starter plan. It is not the same company or product as Bolt.new.

Every number here comes from my own tests and my own monitors. Nothing is estimated or copied, Blink never saw the review first, and every app builder gets the same brief, checks, changes and attacks. This review is based on eight graded tests. I left two out. Payments and emails need your own keys, and only six of the twelve changes could run before my credit floor.

Blink has a Free plan with 5 credits a day, then $25 Starter, $50 Pro and $200 Max, each with more credits. Prices read on 26 October 2026, monthly view. Billed yearly they are lower.

Blink pricing page in the monthly view: Max at $200, Pro at $50, Starter at $25 and Free at $0 a month, with their credits and features listed

Here is the number you will ask about first. In total I paid $25.00, for one month of the Starter plan, and I used 100 of its 105 credits. I cancelled it when the testing was done, so it will not renew for you or me on 26 November 2026.

Blink Usage page: You're on Starter, credits 5.12 left, daily refresh credits reset in 11 hours, plan credits reset in 30 days, and the auto-reload switch turned off

A hosting dialog had an auto-reload switch already on, which would charge $40 to a card when the balance runs low. I turned it off before I published. The twelve changes did not fit in the allowance, so you would need a top-up to finish them. I did not test what 1,000 bookings a month would cost.

Blink's first answer took 12 minutes and 15 credits. It built the member side that you would use as a customer: three classes, sign-up, log-in, password reset and My bookings. It said staff access and Stripe were not connected, and asked what to do next.

Blink editor with the first build's summary on the left, saying staff workflows and Stripe are not connected, and the Fitting Room home page preview on the right

I chose staff access first. If you do the same, expect this: Blink answered that staff access needs a backend, which is a Pro plan feature, and recommended upgrading. I did not. A third message, asking for the staff screens with no separate backend, produced them in 8 minutes for 20 credits. The app was live 31 minutes after the brief, and 20 of the 24 checks had passed by about 75 minutes, in three messages.

A 4, not a 5: three messages and under 90 minutes, but the app was live at 31 minutes and not yet complete.

Mostly. 22 of 24 checks passed. Six more, the payments and the two emails, need keys that are mine to supply, so I left them out. You can book, see My bookings, cancel and hit the full-class rule, and the instructor's attendance list, the CSV and the Ask the studio form all worked for me.

Blink test app member timetable on Saturday 31 October showing Power Reformer with 1 of 4 places taken, 3 spots left, coach Lena Park and a Filter by coach box

Two checks failed. The admin's Edit class button does nothing: Blink refuses the save and shows no message. And the owner's name, Dana Okafor, is nowhere on the site, only "Dana". The pages also add marketing lines the brief never asked for. Nothing was invented that changes a price, a class or an instructor.

I ran six of the twelve changes, one message each, and stopped when the credits fell under my floor of 15. All six worked first time on the live app: a phone field, the role renamed Coach, dark green and cream, an About page, a capacity of 4 for Power Reformer and a coach filter.

Blink test app member timetable on Tuesday 27 October with the Filter by coach box set to Tomás Reyes and only the Mat Flow class listed

Each change cost between 5 and 16 credits. Because six did not run, this test is not graded.

Publishing was quick. The first time I pressed Publish, Blink offered a hosting plan that is free for the first month, and the app answered on its own address about 12 seconds after I pressed Deploy Now. Later publishes took 15 to 20 seconds and cost no credits. A custom domain is listed on Starter, but I did not try one, so I cannot tell you its cost.

Blink dialog reading Your app is live now, with the address fittingroo-app-t94ibxtv.blinkpowered.com, share buttons and an Open App button

The pages carry a "Made with Blink" badge, although the pricing page lists badge removal on Starter. I did not count layout faults or try a real phone.

Every day I also run the Blink.new speed test and the Blink.new security checks.

A review tells you what happened on one day. This tells you what happens every day after. My own monitor opens the app every five minutes for 30 days, from 26 October to 25 November 2026, and records whether it answered and how fast.

Blink hosts the app. I cancelled the plan when the testing ended. Blink says the app stays online until the plan ends on 26 November 2026 and is then paused, so the record should show it going offline after the 30 days.

Every check so far is on the Blink.new uptime record.

Both of my attacks were blocked. Asking for another member's bookings with a member's session returned an empty list, and a delete and a rename on the classes table changed nothing. Writes to the admin table and to my own credits were refused. Password reset by email worked, and sign-up has no role choice.

Blink test app Studio team page for the test admin account, saying Your studio access is not set up yet and showing an account ID to give to the studio admin

Staff access is a row the owner adds, so a visitor cannot make themselves an admin. The reset link opens a page on blink.new, not on the app's own address, which can look odd to your members, and you cannot change it.

Yes: five of six data checks passed. Bookings and the Ask the studio form wrote records, and they stayed after reloads, six redeploys and the colour change. The admin could view, edit and delete rows and export the bookings as a CSV. Filtering rows in the database panel did not work in my run.

Blink editor database panel listing 15 tables and the studio_classes table with its rows: Reformer Basics, Mat Flow, Power Reformer and a probe class

The panel is a real editor, so you can fix a row by hand, which helps when you hit a gap. I used it to type in the 5 credits each member needed, because the pass checkout is not connected.

I did not run payments, because Stripe needs a key that is yours to supply. The app says so itself: pressing Get a pass answers that Stripe checkout is not connected and no charge was made. The six checks that depend on it are left out, and this test is not graded.

Blink test app timetable for a member with no credits, with a notice reading Stripe checkout is not connected yet. No charge was made

Blink sent a verification email and the reset email from its own address with no set-up. A member with no credit is told a class credit is needed.

Exporting means downloading your app so it can run somewhere else. On Starter the code download costs 5 credits, and it is free on Max. I paid it and built the ZIP in an empty folder.

Blink Download Code dialog: Download all project files as a ZIP archive, credits required 5, with an offer to download for free on Max and a Confirm (5 credits) button

The ZIP holds the front end only. There is no database structure and no data, and the code still points at the Blink project for the classes and accounts. The install failed on a peer-dependency conflict, and the build failed until I added React by hand. I did not read the terms, so I cannot say who owns the app.

I sent one question through the message form in the editor: how to stop visitors becoming admin, and how to export the database. A person from Blink replied 4 minutes later. The advice on admin rights was sound. The export answer was open: they needed to confirm it for Starter, and suggested the Pro or Max plan.

Blink workspace with the support message form open on the right showing the end of a reply from a Blink support team member

You cancel in Settings, then Billing, then Cancel Subscription. It takes seven screens, including two that need an answer and a 30% discount offer. The plan page then said the subscription ends on 26 November 2026, and the app is paused after that. I could not see a final invoice yet.

I have not yet read what other users say about Blink.new, so this draft has no review scores. Read them before you decide, and tell me what you find.

If this were my studio, and you were asking me, I would use Blink to get a fast draft of Fitting Room and keep it there. It built real booking logic, and its data rules held against both of my attacks. But I would fix the Edit class button, and I would plan for the credits.

It suits you if you want a working booking app quickly and are happy for their data and accounts to stay on Blink. It suits you less if the app must be able to leave one day, or if you need staff features without paying for the Pro plan. I could not try payments without your own keys, and I could not run all twelve changes.

The eight graded tests average 3.8 out of 5, and I rate Blink.new 3.8. Security is the highest grade, because both attacks failed, and publishing, code export and cost are the lowest, because of the 68 phone score, the front-end-only download and credits that ran low.

See the Blink.new test results and how I rate.