Can v0 build native mobile apps?

v0 produces web applications: Next.js, React, TypeScript, Tailwind CSS and shadcn/ui, deployed to Vercel. That is the stack, and it is a very good one.

Vercel's own staff have been clear about the boundary. Asked directly about native app support, one Vercel team member said v0 "currently focuses on web apps using React, Next.js, and related web tech" and that "something like a PWA would be realistic, but not native apps at the moment." Another said "I'm not sure if v0 has the support for native apps right now." Community threads confirm there is no React Native support either, through Expo or otherwise.

So the answer is not just "not Swift." It is not React Native either. v0 makes web apps, and a web app on a phone is a website.

There is a detail here that is more interesting than it is embarrassing: Vercel built v0's own iPhone app in React Native with Expo — not with v0. Their engineering blog says so openly: "As a company focused on the web, building a native app was new territory for us... we landed on React Native with Expo." When the team behind the tool needed a native app, they reached for something else. That is not hypocrisy, it is just an accurate statement of what v0 is scoped to do.

The usual next idea is to wrap the v0 output — Capacitor, Cordova, or a plain WebView shell — and submit that to the App Store.

Sometimes this works. Often it does not. Apple's review guidelines take a dim view of apps that are essentially a repackaged website, and rejections on those grounds are common enough that you should plan for one rather than hope against it. Even when a wrapper is approved, you inherit the things that make wrapped web apps feel wrong on a phone: scroll physics that are not quite right, keyboard handling that fights the OS, no real access to the platform surface, and a launch experience that reads as a website in a box.

A progressive web app is the more honest version of the same idea, and Vercel's own guidance points there. If your users will genuinely accept installing from a browser and you do not need the App Store, that is a legitimate path. If you need to be in the App Store, it is not.

It is worth being fair here, because if you are reading this you probably like v0, and you are probably right to.

v0 is the fastest way we know to get from an idea to a good-looking, working web interface. The design quality it produces with very little direction is unusual — people routinely get something presentable on the first prompt. The output is clean React that an engineer can pick up and extend rather than throw away. And the ecosystem around it — shadcn/ui, Tailwind, Next.js, instant deployment — means the thing you generate is immediately real.

For landing pages, dashboards, internal tools, marketing sites and web product prototypes, it is genuinely hard to beat. Vercel also states plainly that "Vercel doesn't own the code generated based on your queries and prompts," which is a cleaner ownership statement than several tools in this category manage.

None of that changes what it is scoped to build.

The good news is that a v0 prototype is not wasted work. It is a specification — probably a better one than you would have written in prose.

You have already made the decisions that are hard to make in the abstract: what the screens are, what is on each one, how navigation works, what the states and edge cases look like, what the visual language is. That is most of a product definition, in a form you can point at.

Modaal takes exactly that kind of input. You can start a project from a description, from screenshots, or from design references, and with Plan mode on the agent first researches the architectural approach, confirms the key decisions with you, and writes a PRD documenting what the app does and what it is deliberately deferring. Then it scaffolds the project and implements against that spec — in real Swift for iOS and real Kotlin for Android.

Concretely: screenshot your v0 screens, hand them over, describe the behaviour, and you get a native app rather than a wrapped one. The React does not carry over — nothing about a Next.js component translates to SwiftUI — but the thinking does, and the thinking was the expensive part.

Output. v0 produces React and Next.js for the web. Modaal produces Swift and SwiftUI for iOS, and Kotlin for Android, in a real Xcode project on your own disk.

App Store. A v0 app is not App Store shippable as a native app; the wrapper route risks rejection. Modaal output is a native app you submit through Xcode and App Store Connect with your own developer account.

Pricing model. v0 is credit-based on top of a per-seat subscription — free at $0 with a small monthly credit allowance and a daily message cap, Plus at $30 per user per month, Business at $100 per user per month, which notably carries the same credit allowance as Plus and buys governance rather than capacity. Modaal has a free plan with no prompt cap at all, and Pro is €9 per month billed annually, or €15 monthly, with no credit meter — because inference runs on an AI subscription you already have — 13 agents are supported, including Claude Code, Codex, Cursor, Gemini and Copilot.

Requirements. v0 runs in a browser, anywhere. Modaal needs a Mac with Xcode 26 or newer and around 50 GB free, because it builds through Xcode. If you do not have a Mac, Modaal is not the answer either — a cloud-based native builder would be.

They are not really competitors. Plenty of teams should use both: v0 for the marketing site and the web dashboard, something native for the app. Picking one tool for both jobs is usually the mistake.