Cloudflare unveils new AI tools to boost agentic development

Cloudflare has announced several new offerings to facilitate the development of AI agents, addressing challenges related to complexity, cost, and scalability.

The company's recent developments include partnerships with Auth0 and Stytch to simplify agentic deployment through the Model Context Protocol (MCP). This partnership facilitates delegation of permissions, thus smoothing the deployment of agents. Additionally, Cloudflare has acquired Outerbase to enhance database experiences for developers, streamlining database management.

Cloudflare now offers a free tier for Durable Objects, allowing AI agents to maintain memory and context across interactions. Durable Objects enable more sophisticated actions by providing a foundation for such AI agents. Additionally, the availability of Workflows for building multi-step applications is expected to aid developers in designing applications that can retry, persist, and operate over varying durations.

Matthew Prince, Co-Founder and CEO of Cloudflare, commented, "Cloudflare is the best place to build and scale AI agents. Period. The most innovative companies out there see that agents are the next big frontier in AI, and they're choosing Cloudflare because we've got everything they need to move fast and build at scale on our Workers platform. Cloudflare was built for this moment. First, we built the most interconnected network on the planet. Then, we built a developer platform that took advantage of that network to run code within 50 milliseconds of 95% of everyone online. And, we're keeping our foot on the gas to give developers the best tools to build the future of agentic AI."

The introduction of the remote MCP server is one of the significant advancements. It allows AI agents to interact directly with external services instead of merely issuing instructions. This development is poised to enable agents to perform tasks on a user's behalf, such as sending emails or booking meetings without the need for locally hosted servers. With this, Cloudflare facilitates secure connections over the Internet to services like email through remote MCP servers.

Durable Objects, now accessible on Cloudflare's free tier, expand the capacity for building contextually aware AI agents. These objects combine compute with storage and aid in maintaining statefulness across interactions, ensuring low-latency responses through Cloudflare's expansive network. They enable AI agents to remember previous interactions to provide more tailored responses.

General availability of Workflows further benefits developers by offering a structured approach to multi-step application deployment. Examples include booking processes which require tasks like searching and confirming flight details in a seamless, automated fashion.

Cloudflare's approach to serverless AI deployment allows cost-efficiency through its pay-as-you-go model. By automatically scaling based on demand, Cloudflare's platform promises to reduce costs compared to traditional infrastructures which require fixed capacity provisioning.

Kate Holterhoff, Senior Analyst at RedMonk, remarked, "Cloudflare offers a developer-friendly ecosystem for creating AI agents that includes a free tier for Durable Objects and serverless options for AI inference. These low-cost, easy-to-use options could empower more businesses to adopt and experiment with agentic AI."

Cloudflare continues to expand its infrastructure, striving to make AI inference accessible by deploying GPUs across over 190 cities globally, ensuring AI operations are conducted with minimal latency close to the end user.