July 2025 Summaries

Launched in April, the Vercel Open Source Program aims to provide resources, credits, and support to developers, enabling them to scale their projects confidently. Following the spring 2025 cohort, the program has announced its summer cohort, featuring 28 diverse open-source projects ranging from AI-powered tools and UI component libraries to productivity apps and backend frameworks. Notable projects include OpenCut, a browser-based video editor; Inbox Zero, an AI assistant for email management; and Mobx, a widely used JavaScript state manager. Each project, created by developers from various backgrounds, showcases innovation and community engagement, with projects like Electerm, Resume Matcher, and Taskmaster gaining significant attention on GitHub. The initiative supports the growth and visibility of these projects, encouraging contributions to the open-source community, with applications for the winter cohort opening in January.

The AI SDK 5 is a comprehensive update to the open-source AI application toolkit, enhancing its capabilities for TypeScript and JavaScript developers by introducing type-safe chat integration for major frameworks like React, Svelte, Vue, and Angular. This version focuses on improving the usability and functionality of AI-driven applications with features like redesigned chat, agentic loop control, and speech generation. The SDK emphasizes end-to-end type safety, allowing developers to work more efficiently with language models by distinguishing between UI messages and model messages, ensuring data integrity and ease of persistence. It also introduces flexible and dynamic tools, provider-executed functions, and lifecycle hooks for a seamless AI integration experience. With native support for Server-Sent Events (SSE) for robust data streaming and modular architecture for extensibility, AI SDK 5 caters to the fast-evolving needs of AI applications, backed by a strong community and comprehensive documentation for smooth migration and implementation.

The v0 Ambassador Program has been launched to recognize and support individuals who are actively building and sharing projects using the v0 platform, encouraging innovation and public showcasing of their work. Ambassadors will receive $100 in monthly v0 credits, early access to new features, exclusive swag, increased visibility for their projects, and potential earnings through referral links. They are expected to create and disseminate content about v0 regularly, engage with their audience on various platforms, and provide feedback on new features. The program, currently in its beta phase with limited spots, seeks applicants with a strong connection to v0 and an ability to engage and grow an audience, with applications being reviewed on a rolling basis.

Vercel's Fluid compute, a serverless computing approach introduced alongside Active CPU pricing, has transformed cost efficiency and resource utilization by minimizing cold starts and reducing expenses by up to 95% for over 45 billion weekly requests. Originating from the React team's development of React Server Components and the Next.js team's App Router, Fluid compute required significant infrastructure changes, including a new secure TCP-based protocol for streaming responses from AWS Lambda. This innovation allows multiplexing multiple concurrent requests to the same Lambda instance, breaking away from the traditional one-invocation-per-instance model, thereby optimizing resource usage and performance. The system's Rust-based core adapts to each instance's load profile by gathering metrics to maintain optimal resource utilization, while the new compute-resolver service aids in efficiently routing requests to reuse existing connections. Active CPU pricing further enhances cost savings by charging only for the active CPU time and provisioned memory used, making Vercel an attractive platform for diverse applications, from frontends to AI apps, with Fluid compute now the default for new projects.

The Model Context Protocol (MCP) is a specification designed to standardize the way large language models (LLMs) access external data and systems, extending their capabilities beyond their training data. By acting as a universal connector between AI models and applications, MCP enables developers to expose data sources and tools in a structured manner that AI models can understand, allowing for controlled and predictable interactions. Unlike traditional APIs that require hard-coded calls, MCP uses metadata and structured descriptions to help models decide which tools to use, thus reducing the need for custom integrations for each AI platform. This approach facilitates AI models in performing actions based on real-time data and functions, enhancing their accuracy and utility. MCP servers can be hosted locally or remotely, enabling models to interact with real systems across various domains such as e-commerce, finance, and customer support. While the protocol is open source, created by Anthropic, it is supported by a community of developers and offers a secure way to integrate AI with existing systems. MCP's introduction is comparable to how standards like HTTP and REST revolutionized the web, by providing a consistent framework that reduces integration costs and unlocks new application possibilities.

Vercel has formed a partnership with Solara6, a digital agency renowned for creating high-performing ecommerce experiences for major brands like Kate Spade and Coach. The collaboration aims to enhance ecommerce capabilities by leveraging Solara6's expertise in AI-driven development and fast iteration cycles, alongside Vercel's reliable and flexible infrastructure. This partnership promises improvements in developer velocity, operational costs, page load times, conversion rates, and organic traffic for Solara6's clients, particularly during peak traffic periods such as sales events and product launches. By integrating Solara6's optimization techniques with Vercel's features like Previews, Functions, and Web Analytics, the partnership ensures efficient delivery and robust post-launch monitoring. Joe Krasko, Co-Founder of Solara6, highlights the benefit of an AI-augmented development workflow that enables faster, impactful changes with reduced rework, aligning with the demands of modern ecommerce for enhanced performance, speed, and user experience.

The v0 Platform API is a beta-stage, text-to-app REST API that allows developers to automate the building of web applications by integrating code generation into their existing systems, leveraging v0's infrastructure. It provides a composable interface for generating full-stack web apps from natural language prompts, creating Vercel projects, and managing deployments, all backed by a v0.dev chat link. The API's capabilities include starting chats with custom context, such as files from source code or Git, and using endpoints for natural language app generation. It is accessible through a TypeScript SDK, which simplifies interaction with the API, allowing developers to integrate AI-powered code generation seamlessly. The API is being used to create various applications, including website builders, bots, plugins, and tools that convert natural language prompts into live applications, with functionalities extending to generating UI components and live prototypes. The v0 Platform API template, available as a Next.js application, offers a demonstration of these features, including real-time app generation and project management.

Deployments on Vercel can be secured using Password protection and SSO protection, with both methods applicable to Preview and Production deployments. Password protection is accessible to Teams on Pro and Enterprise plans, while SSO protection is exclusive to the Enterprise plan. Additionally, Vercel Authentication allows team members with Personal Accounts to access deployments using their login credentials. Both Password protection and Vercel Authentication can be enabled simultaneously, offering users the choice of access method. These security features are managed through the Teams Project dashboard, and detailed documentation is available for further guidance.

Grep has integrated support for the Model Context Protocol (MCP), allowing AI applications to query a vast number of public GitHub repositories through a standardized interface. This development enables AI agents to perform search queries and retrieve code snippets from repositories, filtered by language, repository, and file path, with results typically delivered in milliseconds. The Grep MCP server provides an endpoint for these searches, leveraging the same infrastructure as grep.app, and offers ranked snippets based on relevance. Setting up MCP servers is straightforward, as each tool is defined in a machine-readable schema that facilitates predictable integration for agents and applications. For instance, when implementing an MCP server, developers can use Grep's MCP server to find and handle error messages effectively by using specific query patterns. The server's development was streamlined using the mcp-handler package, which adapted Grep's existing API into an MCP server, simplifying both development and deployment.

Grep has enhanced its capabilities by supporting the Model Context Protocol (MCP), enabling AI applications to query vast amounts of code across public GitHub repositories through a standardized interface. This integration allows AI agents to perform search queries and retrieve relevant code snippets based on specific patterns, languages, or repository paths, with results returned almost instantaneously. The Grep MCP server simplifies the process of connecting AI clients by providing a predictable, machine-readable schema for integration, making it straightforward for developers to set up and use. An example illustrates how AI agents can leverage the Grep MCP server to handle error messages effectively by running code searches and refining queries to find optimal solutions. Building on this, the ease of transforming existing APIs into MCP-compliant servers is highlighted, showcasing the simplicity of development and deployment using tools like Vercel's MCP adapter.

Vercel has introduced the AI Cloud, a platform designed to enable developers to build AI-driven applications with ease, speed, and security by leveraging its proven infrastructure capabilities from the Frontend Cloud. The AI Cloud incorporates new tools and infrastructure elements such as AI SDK, AI Gateway, and Fluid compute to simplify and optimize the development and deployment of intelligent applications. These tools facilitate the integration of various AI models, manage computational resources efficiently, and provide secure environments for running untrusted code. Additionally, the platform supports autonomous workflows and offers observability features to monitor and optimize agentic applications. By addressing the unique needs of AI workloads, the AI Cloud aims to transform how developers create and manage applications, ushering in a new era of generative and agentic software development across industries.

NuxtLabs, the creators of the popular fullstack web framework Nuxt and the server runtime Nitro, are joining Vercel, marking a significant milestone for both companies and the future of the open web. This acquisition allows NuxtLabs to continue developing their projects with the same focus and care, maintaining their status as open-source initiatives under an MIT license while keeping the community at the core. Sebastien Chopin, CEO at NuxtLabs, emphasizes that joining Vercel aligns with their values and enhances their ability to support developers without compromising the essence of Nuxt. Vercel is committed to advancing the open web by supporting a variety of frameworks and tools, ensuring that projects like Nuxt and Nitro remain independent and freely available. The collaboration promises further investments in the open-source ecosystem, with upcoming releases such as Nuxt Studio MDC, Nuxt UI Pro, and NuxtHub Admin becoming free and open-source, underscoring a shared vision of supporting developers worldwide.