title: "OpenAI ships Agents API with computer-use in public beta" slug: "openai-ships-agents-api-with-computer-use-in-public-beta" published: "2026-10-17" beat: "Launches" tags: ["Launches", "Tools"] creator: "Agentry Newsroom" editor: "Susanne Sperling, Editor — Human in the Loop" tools: ["Claude (Anthropic)", "Perplexity Sonar"] creativeWorkStatus: "verified" dateReviewed: "2026-10-17" aiActArticle50: "compliant" humanView: "https://agentry.news/launches/openai-ships-agents-api-with-computer-use-in-public-beta" agentView: "https://agentry.news/agent/openai-ships-agents-api-with-computer-use-in-public-beta"
OpenAI released its Agents API in public beta on September 10, 2026, expanding the toolset on September 29 at DevDay to include computer-use capabilities that enable developers to build agents capable
Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. AI policy.
OpenAI introduced its Agents API in public beta on September 10, 2026, providing developers with infrastructure to build and deploy autonomous agents. The release was subsequently expanded at OpenAI's DevDay on September 29, 2026, when the company added computer-use capabilities to the platform OpenAI DevDay 2026 recap.
The computer-use feature allows agents to "interact with software to complete tasks," according to OpenAI's official DevDay announcement OpenAI DevDay 2026 recap. This capability extends beyond text-based interactions, enabling agents to navigate graphical user interfaces and operate applications as a human user would.
OpenAI manages the underlying infrastructure for the Agents API, handling the backend systems that support agent deployment and scaling OpenAI DevDay 2026 recap. The platform ships with an OpenAI-managed Codex agent capable of multi-step work and session continuity, allowing agents to maintain context across multiple interactions and complete complex workflows help.openai.com.
The Agents API release targets developers building enterprise applications that require autonomous task completion. By providing both the base API and a reference implementation through the Codex agent, OpenAI is lowering the barrier to entry for developers experimenting with agentic workflows. The public beta status indicates the platform is stable enough for production use while remaining open to feedback and iteration.
The timing of the computer-use expansion—three weeks after the initial API launch—suggests OpenAI prioritized getting core infrastructure into developer hands before adding the more complex graphical-interaction layer. This phased approach allows developers to begin integrating agents into workflows immediately while preparing for more sophisticated use cases enabled by computer-use capabilities.
The Agents API joins a growing ecosystem of developer tools for building autonomous systems, competing with frameworks and SDKs from other AI labs. OpenAI's managed infrastructure approach differs from open-source alternatives, trading developer control for operational simplicity and guaranteed compatibility with OpenAI's models.