News
Vercel, the company behind Frontend Cloud, has officially launched the AI Cloud, a unified platform designed to support the next wave of AI-native and agentic applications.
Unveiled at Vercel Ship 2025 on July 10, the AI Cloud introduces a set of purpose-built tools and infrastructure for developers building applications powered by large language models, autonomous agents and intelligent workflows. It’s designed to help teams ship conversational interfaces and end-to-end agent systems with speed, scalability and security — without the typical infrastructure overhead.
“The same principles and ease of use you expect from Vercel, now for your agentic applications,” wrote Dan Fein, senior technical product marketing manager at Vercel.
AI Infrastructure for the Agentic Era
"The web is at the early stages of this major shift: building on the decades-long move from purely static to highly dynamic, it is now entering the generative, agentic era."
- Dan Fein
Senior Technical PMM, Vercel
The AI Cloud builds on Vercel’s Frontend Cloud foundation, but adds AI-first capabilities including:
- AI SDK and AI Gateway, allowing developers to integrate any model or tool with minimal configuration and swap providers without code rewrites.
- Fluid compute, which eliminates cold starts and manual scaling by pricing compute only when CPUs are actively in use — a model tailored for bursty and idle-prone AI workloads.
- Secure Sandboxes, where untrusted agent-generated code runs in isolated, ephemeral environments to protect production systems.
- Vercel Queues, enabling background task orchestration and multi-step tool execution without blocking.
- MCP server support, giving agents reliable access to business logic and complex toolchains.
- Vercel BotID, an invisible CAPTCHA designed to stop automation before it reaches sensitive endpoints such as those that trigger LLM calls.
These tools are optimized for autonomous AI systems that generate, reason and act — not just respond. As Fein explained, “Infrastructure should emerge from code, not manual configuration,” an approach Vercel calls “framework-defined infrastructure.”
From Static to Agentic
With the rise of AI, the web is evolving again — from static pages and dynamic applications to generative and agent-driven systems. The AI Cloud supports this shift by enabling both AI-native startups and enterprises integrating AI into legacy applications to move quickly and securely.
Vercel says the platform is already powering AI teams such as Browserbase and Decagon. Its goal is to simplify everything from model calls and memory to agent tool use and long-term observability.
As developers look to build LLM-powered experiences and agentic workflows, the AI Cloud promises a frictionless path — abstracting infrastructure while maximizing control and visibility.
