Cloudflare Announces "Agent Cloud" to Power Next-Generation Agents
Cloudflare announced "Agent Cloud," expanding features for developers to build, deploy, and scale AI agents affordably with new tools like "Dynamic Workers" and "Artifacts."
📋 Article Processing Timeline
- 📰 Published: April 15, 2026 at 00:00
- 🔍 Collected: April 14, 2026 at 15:31
- 🤖 AI Analyzed: April 19, 2026 at 16:10 (120h 38m after Collected)
(April 13, 2026, 6:00 AM Pacific Time announcement)
Cloudflare, the leading connectivity cloud company, today announced new features and the expansion of "Agent Cloud," an agent-oriented cloud that supports developers in building, deploying, and scaling agents. Consisting of infrastructure, security, and developer tools, "Agent Cloud" aims to transition AI agents from experimental demos on local PCs to robust production workloads running on Cloudflare's global network.
The first generation of AI focused on chatbots, but current AI trends are shifting toward coding agents and autonomous tools capable of reading context, reasoning, and executing multi-step actions. However, existing infrastructure relies on always-on, expensive virtual servers or isolated sandboxes, and lacks the scalability to support an environment where every user or employee runs dozens of personal agents simultaneously. Cloudflare aims to solve the challenges of infrastructure, computing, deployment, and security, allowing developers to focus on developing and building next-generation applications.
Matthew Prince, co-founder and CEO of Cloudflare, stated, "The way software is developed is fundamentally changing, and we are entering a world where agents write and execute code. However, agents need a 'home' that is secure by default, can instantly scale to millions, and persists through long-running tasks. Cloudflare has built that foundation with Workers. Today, we are delivering Cloudflare as the definitive platform for the agentic web."
Rohan Varma, Product Manager for Codex at OpenAI, said, "Cloud agents are rapidly establishing themselves as a foundational element of how work gets done. Teaming up with Cloudflare makes it easier for developers to deploy production-ready agents powered by GPT-5.4 and Codex, and run large-scale enterprise workloads."
Cloudflare's "Agent Cloud" provides developers with comprehensive tools and infrastructure to support next-generation AI agents, realizing the following:
"Dynamic Workers" to efficiently scale agents with a purpose-built compute model
For agents to move beyond early adopters and become ubiquitous for everyone, affordability is essential. Running each agent in a separate container incurs massive costs, so current agentic tools are mostly limited to coding assistants for engineers where cost-effectiveness can be expected. Cloudflare is releasing "Dynamic Workers," an isolate-based runtime designed to run AI-generated code in a secure sandbox environment faster and more efficiently than traditional containers. When an agent needs to call an API, transform data, or execute a code snippet for a tool call, "Dynamic Workers" spin up in milliseconds, execute JavaScript, and immediately disappear. The vast majority of agent tasks require secure isolation at 100x the speed and a fraction of the cost of a container, and "Dynamic Workers" scales this to millions of concurrent executions with zero warm-up.
"Artifacts" to empower agents with Git-compatible storage
As AI agents begin to generate more code than ever before, traditional version control platforms struggle to maintain the scale and uptime required for autonomous workloads. To meet this demand, Cloudflare is releasing "Artifacts," a Git-compatible storage primitive built specifically for the agent-first era. "Artifacts" empowers platforms to build next-generation code and file storage at an unprecedented scale...
Cloudflare, the leading connectivity cloud company, today announced new features and the expansion of "Agent Cloud," an agent-oriented cloud that supports developers in building, deploying, and scaling agents. Consisting of infrastructure, security, and developer tools, "Agent Cloud" aims to transition AI agents from experimental demos on local PCs to robust production workloads running on Cloudflare's global network.
The first generation of AI focused on chatbots, but current AI trends are shifting toward coding agents and autonomous tools capable of reading context, reasoning, and executing multi-step actions. However, existing infrastructure relies on always-on, expensive virtual servers or isolated sandboxes, and lacks the scalability to support an environment where every user or employee runs dozens of personal agents simultaneously. Cloudflare aims to solve the challenges of infrastructure, computing, deployment, and security, allowing developers to focus on developing and building next-generation applications.
Matthew Prince, co-founder and CEO of Cloudflare, stated, "The way software is developed is fundamentally changing, and we are entering a world where agents write and execute code. However, agents need a 'home' that is secure by default, can instantly scale to millions, and persists through long-running tasks. Cloudflare has built that foundation with Workers. Today, we are delivering Cloudflare as the definitive platform for the agentic web."
Rohan Varma, Product Manager for Codex at OpenAI, said, "Cloud agents are rapidly establishing themselves as a foundational element of how work gets done. Teaming up with Cloudflare makes it easier for developers to deploy production-ready agents powered by GPT-5.4 and Codex, and run large-scale enterprise workloads."
Cloudflare's "Agent Cloud" provides developers with comprehensive tools and infrastructure to support next-generation AI agents, realizing the following:
"Dynamic Workers" to efficiently scale agents with a purpose-built compute model
For agents to move beyond early adopters and become ubiquitous for everyone, affordability is essential. Running each agent in a separate container incurs massive costs, so current agentic tools are mostly limited to coding assistants for engineers where cost-effectiveness can be expected. Cloudflare is releasing "Dynamic Workers," an isolate-based runtime designed to run AI-generated code in a secure sandbox environment faster and more efficiently than traditional containers. When an agent needs to call an API, transform data, or execute a code snippet for a tool call, "Dynamic Workers" spin up in milliseconds, execute JavaScript, and immediately disappear. The vast majority of agent tasks require secure isolation at 100x the speed and a fraction of the cost of a container, and "Dynamic Workers" scales this to millions of concurrent executions with zero warm-up.
"Artifacts" to empower agents with Git-compatible storage
As AI agents begin to generate more code than ever before, traditional version control platforms struggle to maintain the scale and uptime required for autonomous workloads. To meet this demand, Cloudflare is releasing "Artifacts," a Git-compatible storage primitive built specifically for the agent-first era. "Artifacts" empowers platforms to build next-generation code and file storage at an unprecedented scale...