Cloudflare announced on April 13, 2026, that it is integrating OpenAI’s frontier models, including GPT-5.4, into its Agent Cloud platform. The move allows enterprise customers to deploy AI agents capable of performing autonomous tasks directly within their production environments.
Agent Cloud, which operates on Cloudflare’s Workers AI infrastructure, is designed to process AI workloads at the edge. By bringing OpenAI’s models into this framework, the company aims to reduce latency for global applications. Companies can now use the platform to automate customer service responses, update internal systems, and generate complex business reports.
Scaling enterprise AI workflows
Cloudflare CTO Dane Knecht stated that the partnership is designed to bridge the gap between AI intelligence and end-user performance. "By bringing OpenAI’s powerful models directly into the Cloudflare environment, we are collapsing the distance between intelligence and the end user," Knecht said. He added that the infrastructure is built to ensure these applications are both fast and scalable by default.
In addition to the integration of GPT-5.4, Cloudflare has made the Codex harness generally available within its Cloudflare Sandboxes. This environment provides developers with a secure space to build, run, and test AI applications. The company confirmed that Codex support will also be coming to the Workers AI platform in the near future.
OpenAI’s product lead for Codex, Rohan Varma, emphasized the practical utility of this expansion. "Cloud agents are quickly becoming a foundational building block for how work gets done," Varma said. "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."
This update follows a series of strategic moves by OpenAI to embed its technology into the operations of large-scale enterprises. OpenAI currently counts firms such as Accenture, Walmart, Intuit, and Morgan Stanley among its partners. According to OpenAI, its APIs now process more than 15 billion tokens per minute, supporting over one million business customers worldwide.
