Advertise hereReach developers building with AI coding agents. →

Compiled from official documentation, checked

OpenAI Codex vs GitHub Copilot: documented features compared

Short answer

Codex is OpenAI’s coding agent in a CLI, IDE extension, the ChatGPT desktop app and Codex cloud, included in ChatGPT plans. GitHub Copilot runs in VS Code and other IDEs, in Copilot CLI and as a cloud agent on github.com, with a monthly allowance of GitHub AI Credits. Both support MCP and read AGENTS.md; they differ in config files and how usage is metered.

Copilot’s two agents, step by step: Agent mode in your IDE · Coding agent (cloud agent) on GitHub

How this comparison was made: we have not run hands-on tests or benchmarks of either agent. Every row summarises the vendor’s own documentation and links to it. Prices are list prices as published on the date above and can change.
OpenAI CodexGitHub Copilot
Form factor Codex CLI, IDE extension, ChatGPT desktop app, Codex cloud source VS Code agent mode, other IDEs, Copilot CLI, Copilot cloud agent on github.com source
Add an MCP server codex mcp add <name> -- <command>, or --url for Streamable HTTP source VS Code: guided add flow or mcp.json; Copilot CLI: /mcp add or copilot mcp add source
MCP config location ~/.codex/config.toml, or .codex/config.toml in trusted projects source Copilot CLI: ~/.copilot/mcp-config.json, .mcp.json or .github/mcp.json; VS Code also .vscode/mcp.json source
Remote MCP auth Bearer tokens and OAuth (codex mcp login) source OAuth and headers per server, configured in VS Code or Copilot CLI source
Rules / instructions AGENTS.md: global ~/.codex/AGENTS.md plus one file per directory, 32 KiB default cap source .github/copilot-instructions.md, .github/instructions/*.instructions.md and AGENTS.md source
Individual pricing Included in ChatGPT plans: Free $0, Go $8, Plus $20, Pro from $100/mo source Free $0, Pro $10, Pro+ $39, Max $100/mo, each with a monthly AI credit allowance source
Team pricing ChatGPT Business: $25 per user per month billed monthly (2+ users) source Business $19, Enterprise $39 per granted seat per month source
How usage is metered Plan usage limits (5-hour and weekly) shared with ChatGPT Work; extra usage via ChatGPT credits source GitHub AI Credits: Pro 1,500, Pro+ 7,000, Max 20,000 total per month source

Which fits your setup?

Based on documented features, not our own testing. See how SaaS tools are supported in each: Codex tools · Copilot tools

Full pricing details: OpenAI Codex pricing, plans & limits · GitHub Copilot pricing, plans & limits

Related

AGENTS.md vs CLAUDE.md vs Cursor rules

→

How to add an MCP server to Codex

→

How to add an MCP server to GitHub Copilot in VS Code

→

GitHub Copilot CLI yolo mode: --allow-all and safer tool rules

→

Codex sandbox modes and approval policies explained

→

How to install the OpenAI Codex CLI

→

What is AGENTS.md? The open instruction file for coding agents

→

Codex skills: create, install and use SKILL.md skills in Codex

→

GitHub Copilot CLI: install, sign in and start using it

→

GPT-6.1 Terra in Codex and the API: setup, pricing and limits

→

Codex config.toml: location, precedence and the settings to change

→

Codex on Windows: native sandbox setup, WSL and fixes

→

GitHub Copilot coding agent (now cloud agent): how to use it

→

GitHub Copilot agent mode: how to use it in your IDE

→

GitHub Copilot skills: create, install and use agent skills

→

Claude Code vs OpenAI Codex: documented features compared

→

Cursor vs Claude Code: documented features compared

→

Claude Code vs GitHub Copilot: documented features compared

→

OpenAI Codex vs Cursor: documented features compared

→

Cursor vs GitHub Copilot: documented features compared

→

Google Antigravity vs Claude Code: documented features compared

→

Google Antigravity vs Cursor: documented features compared

→

MCP config generator for every agent

→

Frequently asked questions

We have not tested them hands-on, so we do not rank them. According to their docs, both support MCP and read AGENTS.md. They differ in surfaces, config files and pricing model.

Sources