Claude Code vs Codex for engineering teams.

Claude Code or Codex: which one fits your team

Pick Claude Code if you want Claude: it comes with every paid Claude plan and runs only Anthropic's models (Opus, Sonnet, Fable, Haiku). It works in the terminal, VS Code, JetBrains, a desktop app and the web, and answers @claude on pull requests through GitHub Actions.

Pick Codex if your team already pays for ChatGPT: Plus, Pro and Business include it. It runs OpenAI's GPT models in the CLI, an IDE extension, the ChatGPT desktop app and Codex Cloud, and reviews a pull request when you tag @codex.

For the heaviest users, limits decide it: both meter usage per five hours with weekly limits on top, but Codex's Pro plan has no five-hour limit for now.

Claude Code vs Codex: config files and commands

Claude Code and Codex are built from the same parts, kept in different places. In Claude Code, /import codex copies a Codex setup over: instruction files, MCP servers, subagents and skills.

MCP servers: .mcp.json vs config.toml

Each agent needs its own MCP entry: claude mcp add with --scope project writes JSON to .mcp.json for the whole team, and codex mcp add writes TOML to ~/.codex/config.toml, which the Codex CLI, IDE extension and desktop app share.

Both guard repo config: Claude Code asks before using a checked-in .mcp.json, and Codex reads .codex/config.toml only in a trusted project.

{
  "mcpServers": {
    "shared-server": {
      "type": "http",
      "url": "https://example.com/mcp"
    }
  }
}

[mcp_servers.figma]
url = "https://mcp.figma.com/mcp"
bearer_token_env_var = "FIGMA_OAUTH_TOKEN"

Claude Code vs Codex pricing and limits

Claude Code and Codex, from each vendor's pricing page in October 2026:

• Claude Code comes with Claude Pro at $20 a month ($17 billed yearly), Max from $100 and Team seats at $25 ($20 billed yearly). Every plan's limits reset on a rolling five-hour window, and paid plans add weekly limits on top. More on Claude Code pricing.
• Codex comes with ChatGPT Plus at $20 a month, Pro from $100 and Business at $25 per user ($20 billed yearly), and Free and Go ($8) run GPT-6 Luna in the desktop app. On Plus and Standard Business, OpenAI estimates 5–45 local messages per five hours on GPT-6 Astra, 15–160 on GPT-6.1 Sol and 350–3,000 on GPT-6 Luna. More on Codex pricing.

What our team saw running both

Of the 2,620 pull requests merged in our two repos from July 5 to October 3, 2026, 133 (5%) came from Codex, from 4 engineers. In the 30 days to October 3 the Codex share rose to 9.9%.

The agents' session-analytics blocks put all 31 of Codex's in September, and a little smaller than Claude Code's: a median 576 lines changed to 691.

Limits bit on both: on one engineer's laptop, Claude Code showed a usage-limit notice on 16 of 29 active days from September 4 to October 3, 2026, and in September Codex's weekly window reached 100% on 7 days.

Is Codex or Claude Code better? What our AI reviewer found

Our AI reviewer, IonWarp, reviews every pull request in our two repos and flags should-fix and blocking issues. Comparing Codex's pull requests with Claude Code's from the same engineer, repo and week (August 12 to October 4, 2026), Codex's were flagged 1 point more often, with a 95% confidence interval from 16 fewer to 20 more: no measurable difference.

Pull request size mattered far more. After review, Codex declined 22 of the 58 findings it answered (38%), against 18% for Claude Code with the same engineers (August 31 to September 20); with 31 Codex pull requests, that is thin.

One small team, agent-reported names and a reviewer that changed week to week: how we collected it.

Running Claude Code and Codex on one repository

Keep one instruction file, AGENTS.md. Codex reads it from ~/.codex down to the working folder; Claude Code reads it when the repo has no CLAUDE.md. For Claude-only notes, start a CLAUDE.md with @AGENTS.md, as Anthropic's docs show below, and Claude Code loads both.

@AGENTS.md

## Claude Code

Use plan mode for changes under `src/billing/`.