Recently, we shipped an optimization to our global routing service that reduced its memory usage by 15%, improved time-to-first-byte (TTFB) from the 75th percentile and above by 10%, and significantly improved routing speeds for websites with many static paths. A small number of websites, with hundreds of thousands of static paths, were creating a bottleneck that slowed down our entire routing service. By replacing a slow JSON parsing operation with a Bloom filter, we brought path lookup latency down to nearly zero and improved performance for everyone.

Cold starts have long been the Achilles’ heel of traditional serverless. It’s not just the delay itself, but when the delay happens. Cold starts happen when someone new discovers your app, when traffic is just starting to pick up, or during those critical first interactions that shape whether people stick around or convert. Traditional serverless platforms shut down inactive instances after a few minutes to save costs. But then when traffic returns, users are met with slow load times while new instances spin up. This is where developers would normally have to make a choice. Save money at the expense of unpredictable performance, or pay for always-on servers that increase costs and slow down scalability. But what if you didn't have to choose? That’s why we built a better way. Powered by Fluid compute, Vercel delivers zero cold starts for 99.37% of all requests. Fewer than one request in a hundred will ever experience a cold start. If they do happen, they are faster and shorter-lived than on a traditional serverless platform. Through a combination of platform-level optimizations, we've made cold starts a solved problem in practice. What follows is how that’s possible and why it works at every scale.

In February 2025, Andrej Karpathy introduced the term vibe coding: a new way of coding with AI, “[where] you fully give in to the vibes, embrace exponentials, and forget that the code even exists.” Just months later, vibe coding is completely reshaping how developers and non-developers work. Over 90% of U.S. developers use AI coding tools, adoption is accelerating for other roles, and English has become the fastest growing programming language in the world. We explore this shift in detail in our new State of Vibe Coding. Here are a few of the key takeaways.

Model Context Protocol (MCP) is emerging as a standard protocol for federating tool calls between agents. Enterprises are starting to adopt MCP as a type of microservice architecture for teams to reuse each other's tools across different AI applications. But there are real risks with using MCP tools in production agents. Tool names, descriptions, and argument schemas become part of your agent's prompt and can change unexpectedly without warning. This can lead to security, cost, and quality issues even when the upstream MCP server has not been compromised or is not intentionally malicious. We built mcp-to-ai-sdk to reduce these issues. It is a CLI that generates static AI SDK tool definitions from any MCP server. Definitions become part of your codebase, so they only change when you explicitly update them.

Rox is building the next-generation revenue operating system. By deploying intelligent AI agents that can research, prospect, and engage on behalf of sellers, Rox helps enterprises manage and grow revenue faster. From day one, Rox has built their applications on Vercel. With Vercel's infrastructure powering their web applications, Rox ships faster, scales globally, and delivers consistently fast experiences to every customer.

Founded in 1877, Helly Hansen is a global leader in technical apparel, but its digital experience wasn't living up to its legacy. Operating across 38 global markets with multiple brands (including HellyHansen.com, HHWorkwear.com, and Musto.com), the company was being held back by an outdated tech stack that slowed site speeds and frustrated customers. Through an incremental migration to Next.js and Vercel, Helly Hansen improved Core Web Vitals from red to green, increased developer agility, and delivered a record-breaking Black Friday Cyber Monday, building a foundation for future innovation.

Vercel Log Drains are now Vercel Drains. Why? They’re not just for logs anymore, as you can now also export OpenTelemetry traces, Web Analytics events, and Speed Insights metrics. Drains give you a single way to stream observability data out of Vercel and into the systems your team already rely on.

AI agents are improving at handling complex tasks, but a recurring limitation emerges when they need to access paid external services. The current model requires pre-registering with every API, managing keys, and maintaining separate billing relationships. This workflow breaks down if an agent needs to autonomously discover and interact with new services. x402 is an open protocol that addresses this by adding payment directly into HTTP requests. It uses the 402 Payment Required status code to let any API endpoint request payment without prior account setup. We built x402-mcp to integrate x402 payments with Model Context Protocol (MCP) servers and the Vercel AI SDK.

MongoDB Atlas is now available on the Vercel Marketplace. Developers can now provision a fully managed MongoDB database directly from your Vercel dashboard and connect it to your project without leaving the platform. Adding a database to your project typically means managing another account, working through connection setup, and coordinating billing across services. The Vercel Marketplace brings these tools into your existing workflow, so you can focus on building rather than configuring.

We’re updating Vercel’s Pro plan to better align with how modern teams collaborate, how applications consume infrastructure, and how workloads are evolving with AI. Concretely, we’re making the following changes:

When the MCP standard first launched, many teams rushed to ship something. Many servers ended up as thin wrappers around existing APIs with minimal changes. A quick way to say "we support MCP". At the time, this made sense. MCP was new, teams wanted to get something out quickly, and the obvious approach was mirroring existing API structures. Why reinvent when you could repackage? But the problem with this approach is LLMs don’t work like developers. They don’t reuse past code or keep long term state. Each conversation starts fresh. LLMs have to rediscover which tools exist, how to use them, and in what order. With low level API wrappers, this leads to repeated orchestration, inconsistent behavior, and wasted effort as LLMs repeatedly solve the same puzzles. MCP works best when tools handle complete user intentions rather than exposing individual API operations. One tool that deploys a project end-to-end works better than four tools that each handle a piece of the deployment pipeline.

On September 9, 2025, the campaign extended to DuckDB-related packages after the duckdb_admin account was breached. These releases contained the same wallet-drainer malware, confirming this was part of a coordinated effort targeting prominent npm maintainers. While Vercel customers were not impacted by the DuckDB incident, we continue to track activity across the npm ecosystem with our partners to ensure deployments on Vercel remain secure by default.