Codex is an AI coding partner from OpenAI that helps teams and developers write, review, refactor, and ship code faster. Available as a web app, desktop app, IDE extension and CLI, Codex supports multi-agent workflows, automated background tasks, code reviews, and integrations (GitHub, Slack, Linear). It’s built for engineering teams who want to accelerate development, improve review quality, and automate routine engineering work.

Codex — AI Coding Partner from OpenAI

Codex is an AI coding partner that helps you build and ship with AI, powered by ChatGPT. Built for real engineering work, Codex can tackle tasks from routine pull requests to your most challenging problems—facilitating feature development, migrations, refactors, and more.

Key capabilities

• Built for multi-agent workflows
• A command-center for agentic coding with built-in worktrees and cloud environments
• Agents work in parallel across projects, turning weeks of work into days

• End-to-end task execution, designed to accelerate engineering momentum

• Adapts to how your team works

• Skills expand Codex beyond writing code to code understanding, prototyping, and documentation

• Keeps outputs aligned with your team’s standards and workflows

• Always-on background work

• Automations handle routine yet important tasks like issue triage, alert monitoring, and CI/CD

• Lets your engineers stay focused on higher-leverage work

• The same agent everywhere

• One consistent Codex experience across web, CLI, IDE extension, and iOS

• Cloud-based features like code review and Slack integration to streamline collaboration

• Start fast, move smoothly to your editor

• Begin in the Codex app, then move to your code editor or terminal

• Seamless transitions—from cloud environments to local development, to the terminal

• A productive ecosystem

• Cloud integrations and collaboration tools help teams review, automate, and scale
• Access to the latest models (GPT-5.5, GPT-5.4, GPT-5.3-Codex-Spark) with specialized variants like GPT-5.4-mini

Why Codex helps teams

• Improves quality and speed
• Codex raises baseline design quality, accelerates testing, and strengthens code review

• Teams ship with greater confidence and fewer late-stage surprises

• Scales with your organization

• Plans and credits support individuals, teams, and enterprises

• Flexible pricing with seats, shared workspaces, and enterprise-grade controls

• Enables practical use cases

• Feature development, migrations, and large-scale refactors
• Prototyping, documentation, and code understanding as integral parts of the workflow
• Automatic reviews, cross-tool integrations, and automated background tasks

Pricing and usage at a glance

• Plans range from Free to Enterprise, with Go, Plus, Pro, Business, and Education options
• Usage is token/credit based, not just per message
• The latest models (including GPT-5.5, GPT-5.4, and GPT-5.3-Codex) are available, with variants like GPT-5.4-mini for higher local usage limits
• Pro and higher tiers include promotions that boost Codex usage for a period
• Image generation and other features count toward Codex usage limits; credits can be purchased to extend work without upgrading plans
• For teams, there are credits and workspace controls, plus admin features (SAML SSO, MFA, RBAC) in higher plans

If you hit usage limits, you can purchase extra credits or switch to smaller models to extend your budget. You can also run tasks locally using an API key with standard pricing.