AgentsThe story, in brief

OpenAI's new Agents API gives developers the infrastructure behind Codex and ChatGPT

OpenAI's Agents API is now in public beta. Developers can build autonomous agents that run for hours, execute code, and spawn sub-agents—with no markup beyond token costs.

Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI agents and the coordination of work.AI illustration by KeyNews
The KeyNews take

Why it matters

This is infrastructure democratization: the agent-building primitives that powered Codex and ChatGPT internals are now available to any developer. Practitioners need to evaluate whether to build agents on this platform or alternatives; enthusiasts see OpenAI establishing the baseline API for the agentic era.

The key facts

7 to know
  1. Agents API in public beta

  2. Supports autonomous execution for hours

  3. Code execution capability included

  4. Sub-agent handoff supported

  5. Pricing: token usage only, no platform markup

  6. Sandbox partners: Cloudflare, Vercel, Oracle

  7. Based on infrastructure from Codex and ChatGPT

Go to the source

The Decoderthe-decoder.com

Publisher excerpt: OpenAI is releasing the Agents API as a public beta. It lets developers build cloud agents that run autonomously for hours, execute code, and hand off tasks to sub-agents. There are no extra fees beyond token usage. Cloudflare, Vercel, and Oracle offer additional sandbox environments.
Read original report
Back to today's editionMore agents news

Keep reading

Related stories

More from Agents