Glossary

A full glossary of terms used in Vercel's products and documentation.

A pricing model for Fluid Compute where you only pay for the actual CPU time your functions use while executing, rather than provisioned capacity.

A proxy service from Vercel that routes model requests to various AI providers, offering a unified API, budget management, usage monitoring, load balancing, and fallback capabilities. Available in beta.

A TypeScript toolkit designed to help developers build AI-powered applications with React, Next.js, Vue, Svelte, and Node.js by providing unified APIs for multiple LLM providers.

See Web Analytics.

A network topology that shares an IP address among multiple nodes, routing requests to the nearest available node based on network conditions to improve performance and fault tolerance.

The process that Vercel performs every time you deploy your code, compiling, bundling, and optimizing your application so it's ready to serve to users.

A cache that stores build artifacts and dependencies to speed up subsequent deployments. Each build cache can be up to 1 GB and is retained for one month.

The command used to build your project during deployment. Vercel automatically configures this based on your framework, but it can be overridden.

A file-system-based specification for a directory structure that can produce a Vercel deployment, primarily targeted at framework authors.

Security features that help identify and block malicious bots and crawlers from accessing your applications.

A distributed network of servers that stores static content in multiple locations around the globe to serve content from the closest server to users.

Development practices where code changes are automatically built, tested, and deployed. Vercel provides built-in CI/CD through Git integrations.

The Vercel CLI is a command-line tool that allows you to deploy projects, manage deployments, and configure Vercel from your terminal.

The processing power and execution environment where your application code runs. Vercel offers serverless compute through Functions and Edge compute through Middleware.

The ability to handle multiple requests simultaneously. Vercel Functions support concurrency scaling and Fluid Compute offers enhanced concurrency.

Key metrics defined by Google that assess your web application's loading speed, responsiveness, and visual stability, including LCP, FID, and CLS.

Scheduled tasks that run at specified intervals. Vercel supports cron jobs for automating recurring processes.

A domain that you own and configure to point to your Vercel deployment, replacing the default .vercel.app domain.

A specialized cache that stores responses from data fetches in frameworks like Next.js, allowing for granular caching per fetch rather than per route.

A type of cyber attack where multiple systems flood a target with traffic. Vercel provides built-in DDoS protection and mitigation.

URLs that accept HTTP POST requests to trigger deployments without requiring a new Git commit.

The result of a successful build of your project on Vercel. Each deployment generates a unique URL and represents a specific version of your application.

Security features that restrict access to your deployments using methods like Vercel Authentication, Password Protection, or Trusted IPs.

A file system structure used to organize and store files, also known as a folder. Often abbreviated as "dir" in programming contexts.

The edge refers to servers closest to users in a distributed network. Vercel's Edge Network runs code and serves content from edge locations globally.

A global data store that enables ultra-fast data reads at the edge (typically under 1ms) for configuration data like feature flags.

Vercel's global infrastructure consisting of Points of Presence (PoPs) and compute-capable regions that serves content and runs code close to users.

A minimal JavaScript runtime that exposes Web Standard APIs, used for Vercel Functions and Routing Middleware.

A context for running your application, such as Local Development, Preview, or Production. Each environment can have its own configuration and environment variables.

Configuration values that can be accessed by your application at build time or runtime, used for API keys, database connections, and other sensitive information.

Data transfer between the Vercel Edge Network and user devices, optimized for performance and charged based on usage.

Configuration switches that allow you to enable or disable features in your application without deploying new code, often stored in Edge Config.

See Vercel Firewall.

An enhanced execution model for Vercel Functions that provides in-function concurrency, and a new pricing model where you only pay for the actual CPU time your functions use while executing, rather than provisioned capacity.

A software library that provides a foundation for building applications. Vercel supports over 30 frameworks including Next.js, React, Vue, and Svelte.

A configuration setting that tells Vercel which framework your project uses, enabling automatic optimization and build configuration.

See Vercel Functions.

Automatic connection between your Git repository (GitHub, GitLab, Bitbucket, Azure DevOps) and Vercel for continuous deployment.

HTTP headers that can be configured to modify request and response behavior, improving security, performance, and functionality.

Secure HTTP protocol that encrypts communication between clients and servers. All Vercel deployments automatically use HTTPS with SSL certificates.

Processes limited by input/output operations rather than CPU speed, such as database queries or API requests. Optimized through concurrency.

Automatic optimization of images including format conversion, resizing, and compression to improve performance and reduce bandwidth.

A feature that allows you to update static content without redeployment by rebuilding pages in the background on a specified interval.

The command used to install dependencies before building your project, such as npm install or pnpm install.

Third-party services and tools that connect with Vercel to extend functionality, available through the Vercel Marketplace.

TLS fingerprinting techniques used by Vercel's security systems to identify and restrict malicious traffic patterns.

A feature that allows you to send logs to external services for long-term retention and analysis.

Vercel's fully managed platform that handles server provisioning, scaling, security, and maintenance automatically.

A protocol for AI applications that enables secure and standardized communication between AI models and external data sources.

Code that executes before a request is processed, running at the edge to modify responses, implement authentication, or perform redirects.

A version control strategy where multiple packages or modules are stored in a single repository, facilitating code sharing and collaboration.

A version control strategy where each package or module has its own separate repository, also known as "polyrepo."

Applications that serve multiple customers (tenants) from a single codebase, with each tenant getting their own domain or subdomain.

A JavaScript runtime environment that Vercel supports for Vercel Functions and applications.

Tools and features that help you monitor, analyze, and understand your application's performance, traffic, and behavior in production.

A federation protocol that issues short-lived, non-persistent tokens for secure backend access without storing long-lived credentials.

The server that stores and runs the original version of your application code, where requests are processed when not served from cache.

The folder containing your final build output after the build process completes, such as dist, build, or .next.

A collection of files and directories grouped together for a common purpose, such as libraries, applications, or development tools.

A deployment protection method that restricts access to deployments using a password, available on Enterprise plans or as a Pro add-on.

Distributed servers in Vercel's Edge Network that provide the first point of contact for requests, handling routing, DDoS protection, and SSL termination.

A deployment created from non-production branches that allows you to test changes in a live environment before merging to production.

The live version of your application that serves end users, typically deployed from your main branch.

An application that you have deployed to Vercel, which can have multiple deployments and is connected to a Git repository.

A performance metric in Speed Insights that uses real user data to measure your application's actual performance in production.

HTTP responses that tell clients to make a new request to a different URL, useful for enforcing HTTPS or directing traffic.

Geographic locations where Vercel can run your functions and store data. Vercel has 19 compute-capable regions globally.

A location where files and source code are stored and managed in version control systems like Git, maintaining history of all changes.

URL transformations that change what the server fetches internally without changing the URL visible to the client.

The execution environment for your functions, such as Node.js, Edge Runtime, Python, or other supported runtimes.

Logs generated by your functions during execution, useful for debugging and monitoring application behavior.

An authentication protocol that allows Enterprise teams to log into Vercel using their organization's identity provider.

See Vercel Sandbox.

An Enterprise feature that creates private connections between Vercel Functions and backend infrastructure using dedicated IP addresses.

A cloud computing model where code runs without managing servers, automatically scaling based on demand and charging only for actual usage.

Performance monitoring that provides detailed insights into your website's Core Web Vitals and loading performance metrics.

Vercel's suite of storage products including Blob storage for files and Edge Config for configuration data.

A technique for sending data progressively from functions to improve perceived performance and responsiveness.

A deployment protection method that restricts access to deployments based on IP address allowlists, available on Enterprise plans.

A high-performance build system for monorepos that provides fast incremental builds and remote caching capabilities.

An AI-powered tool that converts natural language descriptions into React code and UI components, integrated with Vercel for deployment.

A deployment protection method that restricts access to team members and authorized users with Vercel accounts.

Scalable object storage service for static assets like images, videos, and files, optimized for global content delivery.

A multi-layered security system that protects applications from threats, including platform-wide DDoS protection and customizable WAF rules.

Serverless compute that allows you to run server-side code without managing servers, automatically scaling based on demand.

An ephemeral compute primitive for safely running untrusted or user-generated code in isolated Linux VMs.

A predictive performance metric that anticipates the impact of changes on application performance before deployment.

A customizable security layer that allows you to define rules to protect against attacks, scrapers, and unwanted traffic.

Privacy-friendly analytics that provide insights into website visitors, page views, and user behavior without using cookies.

In JavaScript, an entity in a repository that can be either a single package or a collection of packages, often at the repository root.