---
title: "OpenAI launches Agents API in public beta"
slug: "openai-launches-agents-api-in-public-beta"
published: "2026-09-21"
beat: "Launches"
tags: ["Launches", "Tools"]
creator: "Agentry Newsroom"
editor: "Susanne Sperling, Editor — Human in the Loop"
tools: ["Claude (Anthropic)", "Perplexity Sonar"]
creativeWorkStatus: "verified"
dateReviewed: "2026-09-21"
aiActArticle50: "compliant"
humanView: "https://agentry.news/launches/openai-launches-agents-api-in-public-beta"
agentView: "https://agentry.news/agent/openai-launches-agents-api-in-public-beta"
---# OpenAI launches Agents API in public beta

> OpenAI released the Agents API in public beta on September 10, 2026, giving all developers access to a managed infrastructure for building autonomous agents with orchestration, sessions, and tool use—

*Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. [AI policy](/ai-policy).*

OpenAI introduced the **Agents API in public beta** on September 10, 2026, opening access to the same orchestration and tool-use harness that powers its Codex infrastructure [OpenAI](https://openai.com/index/introducing-the-agents-api/). The release makes the managed agent framework available to all developers through a single, unified API call.

## What the API delivers

The Agents API provides three core capabilities: **orchestration** to coordinate agent behavior, **sessions** for stateful agent interactions, and **tool use** to let agents call external functions and APIs. Rather than requiring developers to build these primitives from scratch, [OpenAI's announcement](https://openai.com/index/introducing-the-agents-api/) positions the API as a simplified interface to battle-tested infrastructure that has already powered production agent workloads.

Developers can integrate the API into applications that need autonomous decision-making, multi-step reasoning, or integration with external systems. The public beta status means the API is stable and ready for production use, though OpenAI may refine features based on developer feedback.

## Pricing and availability

There are no additional fees for using the Agents API itself [MarketechPost](https://www.marktechpost.com/2026/09/10/openai-launches-the-agents-api-in-public-beta-putting-the-codex-harness-behind-one-api-call/). Developers simply pay for the tokens their agents consume and for any external tools they invoke—a straightforward model that aligns agent economics with actual usage.

The timing of the release reflects OpenAI's shift toward putting agentic capabilities directly into developer hands. Rather than keeping agent orchestration proprietary or experimental, the company is standardizing it as a product that can be embedded across a wide range of applications.

## Market context

The Agents API entry into public beta marks a concrete milestone in the agent economy. Unlike roadmaps or research prototypes, the API is shipping and usable today. This contrasts with earlier periods when agent frameworks were fragmented across competing vendor implementations or required custom engineering.

For enterprises and startups building agent-native applications, the release reduces friction: they no longer need to license multiple tools or maintain custom orchestration logic. The managed nature of the API also means OpenAI handles infrastructure scaling, session persistence, and tool routing—layers that previously required significant engineering effort.

## What comes next

The public beta phase typically runs for weeks to months before a general availability release. During this window, developers will likely report integration patterns, performance findings, and edge cases that inform the final product shape.