v0 by Vercel
Also known as: v0, v0.dev, v0.app, Vercel v0, v0 by Vercel
Vercel's AI app builder that turns natural language into production React and Next.js code, with a full stack sandbox, Git workflows, one click deployment, and enterprise security.
v0 is an AI development tool built by Vercel, the company behind the Next.js framework, that turns natural language into production quality web code. Launched in late 2023 as a text to component experiment, it has grown into one of the most used AI tools in the React ecosystem, with more than six million developers, and it was renamed from v0.dev to v0.app in early 2026. You describe what you want, a pricing page, a dashboard, a multi page application, and v0 generates clean React and Next.js code that follows accessibility and responsive best practices using the shadcn component library, output that experienced engineers are willing to ship.
A major update in early 2026 pushed v0 well beyond components. It added a full stack sandbox that runs Next.js applications with server actions and database work, a VS Code style editor, a Git panel that creates branches and pull requests directly from chat, database connections including Supabase, Snowflake, and Amazon services, and agentic planning. Vercel now describes v0 as an AI agent for building real code, full stack applications, and agents, positioning it for production software rather than throwaway demos. It can import a GitHub repository, pull environment variables, and work on existing applications, not just new ones.
v0 runs its own family of models, Mini, Pro, and Max, each fine tuned for frontend work and selectable per task to trade quality against cost. Deployment is a one click push to Vercel's infrastructure, and all generated code is standard React and Next.js that can be exported to a local machine or GitHub, so teams are not locked in at the code level, though the workflow is tightly coupled to the Vercel ecosystem. A headless Platform API exposes the same prompt to app to deployment lifecycle, and third parties build their own app builders, bots, and agents on top of it.
For organizations, v0 addresses the governance gap that vibe coding created, offering enterprise grade security with SOC 2 compliance, single sign on through SAML, audit logs, data opt out for sensitive work, and service level agreements on higher tiers. The honest limits are that v0 remains frontend first: its interface quality leads the market, but backend, authentication, and database logic still lean on external integrations, and it is strongest inside the Vercel and Next.js world. Pricing runs from a free tier with a small monthly credit allowance through Premium at twenty dollars, Team at thirty dollars per user, Business at one hundred dollars per user, and custom enterprise plans, metered in tokens so heavier generations cost more.
Vendor details
Canonical URL
https://v0.app
Category
Coding agent
Subcategory
AI app builder for React and Next.js (Vercel, full stack sandbox)
Funding status
v0 is a product of Vercel, the independent company founded by Guillermo Rauch, creator of the Next.js framework. Vercel has raised capital from investors including Accel, GV, and Greenoaks Capital, and has been reported at a valuation of around 9.3 billion dollars. v0 launched in October 2023 and, as of 2026, reports more than six million developers on the platform.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Integrates GitHub for repository import, sync, branch creation, and pull requests from chat, and Vercel for one click deployment and environment variable import. Connects to databases and data platforms including Supabase, Snowflake, and Amazon services, imports designs from Figma, and pulls context from source code, Git, the shadcn registry, screenshots, and attachments. A headless Platform API wraps the full prompt to project to code to deployment lifecycle, letting developers embed v0 into their own tools, bots, command line tools, and agents. Deployment targets Vercel's infrastructure, and generated code exports to local or GitHub.
In practice
You need a polished dashboard or landing page fast. Describe it in plain English and v0 generates production React and Next.js code using shadcn components, then deploys it to Vercel with one click.
You want to build a full stack prototype, not just UI. v0's sandbox runs a Next.js app with server actions and a Supabase database, and its Git panel opens pull requests straight from chat.
You are an enterprise worried about ungoverned vibe coding. v0's Business and Enterprise tiers add SOC 2 compliance, single sign on, audit logs, and data opt out so app generation stays inside your controls.
Sources & related URLs
Related / legacy domains
Capability coverage
8.0 / 14 capabilities · 57%
| Integrations & Tool CallingIntegrates GitHub for repository import, sync, branch creation, and pull requests from chat, deploys to Vercel with environment variable import, connects databases and data platforms including Supabase, Snowflake, and Amazon services, imports from Figma and the shadcn registry, and exposes a Platform API, broad named integrations with real action. | Full |
|---|---|
| Workflow OrchestrationAdds agentic planning and, through a full stack sandbox, generates and runs complete Next.js applications from a prompt, opening pull requests and deploying, and Vercel now describes it as an AI agent for building real code, full stack applications, and agents, a genuine end to end build and ship orchestration. | Full |
| Knowledge Grounding & RAGGrounds generation in supplied context, letting a chat start from source code, a Git repository, the shadcn registry, screenshots, or attachments, and can import a GitHub repository to work on an existing application, real context grounding, though based on provided context rather than a deep codebase knowledge graph. | Partial |
| Human Oversight & GuardrailsKeeps humans in control through pull requests, preview deployments, and a design mode for guided edits, and its enterprise release emphasizes an audit trail to curb ungoverned code creation, real review and approval oversight rather than a runtime guardrail enforcement engine. | Partial |
| Security, Identity & GovernanceProvides enterprise grade security on higher tiers, including SOC 2 compliance, single sign on through SAML, audit logs, data opt out for sensitive work, service level agreements, and advanced analytics, a comprehensive security, identity, and governance matrix. | Full |
| Observability & AuditabilityOffers audit logs and advanced analytics on enterprise tiers, plus preview deployments and Git history for visibility into what was built and shipped, real auditability, short of a comprehensive agent execution tracing and metrics suite. | Partial |
| Memory & State PersistencePersists projects and chat history that can be resumed and uses Git branches for versioning, but these are project state and standard version control, not a distinct persistent agent memory or checkpoint capability documented as first class. | Unable to verify |
| Deployment & Data ResidencyRuns as a hosted service that deploys to Vercel, but all generated code is standard React and Next.js that exports fully to a local machine or GitHub with no code level lock in, and a Business tier adds data opt out, real portability and data control, though the tool itself is not self hosted. | Partial |
| Prebuilt Agents, Templates & PacksShips reusable building blocks through the shadcn component registry, design systems, and templates that generations draw on, a real library of components and patterns, though not a browsable marketplace of prebuilt autonomous agents. | Partial |
| Triggers & Channel CoverageReaches users through the web application, an in browser editor, and a headless Platform API that embeds app generation into external tools, bots, command line tools, customer systems, and custom agents, real channel breadth that is prompt and API invoked rather than driven by broad external event triggers. | Partial |
| Model Flexibility & RoutingRuns its own family of models, Mini, Pro, and Max, each fine tuned for frontend code generation and selectable per task to trade quality against cost, genuine own multi variant models with per task routing. | Full |
| APIs, SDKs & MCP ExtensibilityExposes a headless Platform API, a composable REST interface wrapping the full prompt to project to code to deployment lifecycle with bring your own context, that third parties build entire app builders, bots, and agents on, a strong and first class extensibility surface, though a dedicated software development kit and Model Context Protocol server are not clearly documented. | Partial |
| Testing, Debugging & OptimizationGenerates code and runs previews, but does not ship a dedicated testing, debugging, or code quality engine, and no first class capability for generating tests, detecting defects, or evaluating and optimizing code is documented. | Unable to verify |
| Browser & Computer UseRuns a full stack sandbox that executes generated Next.js applications, including server actions and database operations, and produces live preview environments, real code execution within the platform, though not general autonomous browser automation. | Partial |
Pricing
Free tier includes 5 dollars in monthly credits. Premium is 20 dollars a month, Team is 30 dollars per user a month, Business is 100 dollars per user a month, and Enterprise is custom. Generations are metered in tokens by model and complexity.
Flat monthly subscription tiers that each include a credit allowance, with AI generations metered in tokens. Each generation draws down credits based on the model chosen (Mini, Pro, or Max) and the complexity of the prompt, so heavier full stack builds cost more, and Premium and Team users can buy additional credits.
Included quota
Free: five dollars in monthly credits, a standard model, GitHub sync, Vercel deploy, design mode, and a daily message limit. Premium at twenty dollars adds twenty dollars in monthly credits, Figma imports, and API access. Team at thirty dollars per user adds a shared credit pool and collaboration. Business at one hundred dollars per user adds data opt out for sensitive work. Enterprise adds SOC 2, single sign on, audit logs, service level agreements, and advanced analytics. All generations meter tokens by model and complexity, and paid users can buy more credits.
What is public
Public: the five tier prices, the free tier contents, per model token rates, and the enterprise governance features. Not public: exact credit consumption per generation and enterprise dollar pricing.
Billing mechanics
Five flat monthly tiers from free to enterprise, each with a credit allowance, plus token metered generation cost that varies by model and complexity, with extra credits purchasable on paid plans.
Cost watchouts
Token based billing makes cost hard to predict, since a simple component costs pennies but a complex full stack generation can burn a monthly credit allowance in a few prompts. Higher tier models cost far more per token than the mini model. Production hosting on Vercel may require a separate paid Vercel plan beyond the v0 subscription.
Variable cost rationale
Billing is token metered, so cost tracks the model chosen and the complexity of each generation, and a complex full stack build can consume a monthly credit allowance in a few prompts. Even the flat tiers are really credit budgets that heavy use exhausts, prompting credit top ups or a higher tier. Cost therefore scales with usage and is, by common report, hard to predict, so exposure is high.
Additional watchouts
Token metered cost is unpredictable, complex full stack generations burn credits fast, higher tier models are far more expensive per token, and production hosting on Vercel can be an additional charge. v0 is also frontend first, so backend and database work often needs external tools.
Overage / add-ons
Once a plan's included credits are used, Premium and Team users can purchase additional credits. Generation cost scales with the chosen model's per token input and output rates and the size of the prompt and output.
Sales call required
No — self-serve available
Free / trial
A free plan includes five dollars in monthly credits that reset each cycle, access to a standard model, GitHub sync, Vercel deployment, and design mode, with a daily message limit, enough to explore and build simple prototypes without a card.
Lowest paid plan
Premium at twenty dollars a month is the lowest individual paid plan, above a free tier with five dollars in monthly credits.
Commercial notes
v0 sits in the middle of the AI app builder market on price, justified by best in class user interface output, deep Vercel and Next.js integration, and a three model system that lets users trade quality against credit burn. Its enterprise tiers add genuine governance, which most vibe coding rivals lack, but token metering makes budgeting harder than flat rate tools.
Key ambiguities
Because billing is token metered, the real monthly cost depends on model choice, prompt complexity, attachments, and how much full stack work is generated, which is not knowable until a generation runs. Enterprise pricing is custom, and production Vercel hosting may be a separate cost.
Cancellation / refund
Self serve monthly plans can be changed or cancelled by the user, and Premium and Team users can top up credits. Enterprise terms are arranged with Vercel's sales team.
Missing data
Exact credit to token conversion for a given generation, and enterprise dollar pricing, are not published, so real cost per project must be estimated from usage.
Related vendors
- Cognition — Maker of Devin, an autonomous AI software engineer
- Aider — Open-source, model-agnostic terminal coding agent that works…
- Anthropic Claude Code — Anthropic's terminal and IDE-native coding agent with sub-agent…
- Augment Code — AI coding assistant evolving into an AI-native engineering platform
- Blackbox AI — AI coding assistant for code generation and developer workflows
- Blink — Agentic AI app builder that turns a natural language prompt into a…