Claude Code vs Codex
Direct comparison of Claude Code (from Anthropic) and Codex (from OpenAI). Want a 3- or 4-way comparison? Open the multi-agent tool →
| Claude Code Anthropic | Codex OpenAI | |
|---|---|---|
| Overview | ||
| Vendor | Anthropic | OpenAI |
| Category | coding | coding |
| Release date | — | — |
| Source | proprietary | proprietary |
| License | — | — |
| Pricing | Included with Claude Pro/Max/Team subscriptions | Included with ChatGPT Plus/Pro |
| Base models | ||
| Capabilities | terminalfile_iomulti_file_editshell_exec+1 | terminalfile_ioshell_execcloud_runtime |
| Capability | ||
| gpt.buzz Index | 95.4 | 70.0 |
| SWE-bench Verified | 80.9% | 85.0% |
| SWE-bench Pro | 55.4% | 56.8% |
| Terminal-Bench 2.0 | 65.4% | 77.3% |
| GAIA | 74.6% | — |
| WebArena | 64.5% | — |
| Adoption | ||
| OpenRouter · daily | 51.6B | — |
| OpenRouter · cumulative | 3.52T | — |
| GitHub stars | — | — |
| Links | ||
| Links | ||
More on each
Full agent profiles — Claude Code / Codex
Or browse the full agent directory / composite leaderboard.