Build and run long-lived cloud agents with OpenAI's managed Codex harness, tools, environments, and subagents.
OpenAI Agents API is a public-beta API for building production agents on the managed harness and infrastructure behind Codex. It handles long-running sessions, context management, tool use, subagents, and execution environments while letting developers define the task, model, tools, files, and compute location. Agents can use OpenAI-hosted sandboxes, a team's own infrastructure, or supported sandbox providers.
Data is being prepared
Traffic data is refreshed monthly.
Run parallel coding agents from JetBrains in isolated worktrees.
Run software delivery workflows with governed AI agents.
Give coding agents a Mac, browser, files, and durable sessions
Build and launch production AI agents in one platform