← all releases
OpenAI·Sep 10, 2026·Public Beta·paid

Agents API

The Agents API is an OpenAI managed cloud service that allows developers to build and execute long-running AI agents using the Codex harness. It provides context management with automated compaction, multi-agent coordination with subagent support, programmatic tool calling, and flexible deployment options across OpenAI-hosted sandboxes or third-party infrastructure.

featured on blogAI AgentsAgentic AILLMsInfrastructure
visit website docs ↗
on the blogOpenAI Introduces Agents API Powered by Codex Harness 4 min
overview

The Agents API is an OpenAI managed cloud service that allows developers to build and execute long-running AI agents using the Codex harness. It provides context management with automated compaction, multi-agent coordination with subagent support, programmatic tool calling, and flexible deployment options across OpenAI-hosted sandboxes or third-party infrastructure.

key features
  • 01Automates execution, state tracking, and error handling for agent workflows
  • 02Condenses long conversation histories automatically to prevent reaching token limits
  • 03Delegates tasks to parallel subagents with isolated context windows to speed up processing
  • 04Runs tools in parallel and filters data programmatically to minimize context overhead
  • 05Executes code securely within OpenAI-hosted sandboxes or partner compute environments
Ad
Master AI Marketing: Work Smarter, Not Harder

Master AI Marketing: Work Smarter, Not Harder

Unlock the power of AI to automate workflows and scale your results. Join 1.5M+ professionals and stay ahead of the curve with the latest AI insights.

Sponsored
related productsbrowse all →
AIO Sandboxopen sourceAIO Sandbox is an integrated environment designed for AI agents, combining a browser, terminal, filesystem, VSCode, Jupyter, and MCP Server into a single Docker container. This unified setup allows seamless development and execution of AI agents without the need for multiple services.Fractalsopen sourceFractals is an open-source recursive agentic task orchestrator developed by TinyAGI. It enables the decomposition of high-level tasks into self-similar trees of executable subtasks, which are then executed by an agent swarm in isolated git worktrees.llmockopen sourcellmock is a deterministic mock LLM server designed for testing purposes. It provides a real HTTP server with authentic SSE streams, allowing developers to simulate interactions with various LLM APIs like OpenAI, Claude, and Gemini without incurring costs or dealing with non-deterministic results.
Master AI Marketing: Work Smarter, Not Harder

Master AI Marketing: Work Smarter, Not Harder

Unlock the power of AI to automate workflows and scale your results. Join 1.5M+ professionals and stay ahead of the curve with the latest AI insights.

Sponsored