Cloudflare is expanding its Agent Cloud with new features to help developers build, deploy and scale agents. This suite of infrastructure, security, and developer tools is designed to move AI agents from experimental demos on local laptops to robust, production-grade workloads running across Cloudflare’s global network.
While the first generation of AI focused on chatbots, the industry is shifting toward coding agents and autonomous tools that read context, reason, and take multi-step actions. However, existing infrastructure — relying on expensive, always-on virtual servers or isolated sandboxes —cannot scale to a world where every user and employee has dozens of personal agents running simultaneously. Cloudflare aims to address infrastructure, compute, deployment and security so developers can focus on building the next generation of applications.
“The way people build software is fundamentally changing. We are entering a world where agents are the ones writing and executing code,” said Matthew Prince, co-founder and CEO of Cloudflare. “But agents need a home that is secure by default, scales to millions instantly, and persists across long-running tasks. We’ve spent nine years building the foundation for this with Cloudflare Workers. Today, we are making Cloudflare the definitive platform for the agentic web.”
“Cloud agents are quickly becoming a foundational building block for how work gets done, and with Cloudflare, we’re making it dramatically easier for developers to deploy, production-ready agents powered by GPT-5.4 and Codex to run real enterprise workloads at scale,” said Rohan Varma, Product, Codex, at OpenAI.
Cloudflare’s Agent Cloud provides a full suite of tools and infrastructure to power the next generation of AI agents, allowing developers to:
Scale agents efficiently with a purpose-built compute model
For agents to move beyond early adopters and into everyone’s hands, they also have to be affordable. Running each agent in its own container is expensive, so today’s agentic tools are often limited to coding assistants for engineers who can justify the cost. Cloudflare is rolling out Dynamic Workers, an isolate-based runtime designed to run AI-generated code in a secure, sandboxed environment faster and more efficiently than traditional containers. When an agent needs to execute a code snippet to call an API, transform data, or chain tool calls together, Dynamic Workers spin up in milliseconds, run the JavaScript, and disappear. For the vast majority of agent tasks, this is all you need: secure isolation at 100x the speed and a fraction of the cost of containers, scaling to millions of concurrent executions with no warm-up.
Power agents with Git-compatible storage
As AI agents begin to generate more code than ever before, traditional version control platforms are struggling to maintain the scale and uptime required for autonomous workloads. To meet this demand, Cloudflare is introducing Artifacts, a Git-compatible storage primitive built specifically for the agents-first era. Now Cloudflare is providing a powerful foundation that allows platforms to build next-generation code and file storage at an unprecedented scale. Artifacts enables developers to create tens of millions of repositories, fork from any remote source, and provide agents with a permanent home for code and data that is accessible to any standard Git client.
Give agents their own computers with Sandboxes
Cloudflare is announcing the General Availability of Sandboxes for when an agent needs a full operating system. A Sandbox is a persistent, isolated Linux environment with a shell, a filesystem, and background processes, where an agent can clone a repository, install Python packages, run builds, and iterate with the same tight feedback loop a human developer gets.
Build more capable agents with Think
Current agents are limited by their short lifespans, yet the tasks they are meant to solve are often long-lived and cross-platform. Cloudflare’s Agent Cloud addresses this disconnect with Think, a framework within the Agents SDK designed for persistence. This next-generation SDK empowers developers to build agents that support long-running, multi-step tasks rather than just responding to single prompts.
Ensure reliability and flexibility with a unified AI platform
Following up on Cloudflare’s acquisition of Replicate, Cloudflare is expanding its model catalog to allow developers to choose from a range of state-of-the-art proprietary models from OpenAI and GPT-5.4 and open-source models, all through a single pane of glass. With the current pace of change in models, committing to a single provider could mean potentially missing out when another, better model comes along. Cloudflare makes switching between providers as easy as changing a single line of code, without the requirement of managing multiple vendors.
Discussion about this post