OpenAI Codex vs Aider: what developers actually say
Based on 2,047 opinions on Hacker News from developers who discussed these two, most of them first-hand users.
Codex comes out ahead on evidence and approval: 2,013 opinions at 69% approval against 34 opinions at 60% for Aider. Codex code-quality mentions run 343 positive to 116 negative, and it leads on cost/pricing (174 to 54) and autonomy (169 to 46). Its weak spot is reliability at 112 positive and 127 negative mentions, and 90-day approval has fallen to 63. Aider is the better pick for very large codebases, where its mentions run 3 to 1 positive and the one explicit preference cites a repo with thousands of files.
When developers explicitly said they prefer one or switched: 0 picked OpenAI Codex, 1 picked Aider.
Choose OpenAI Codex if
- Want a large track record to judge by
- Value autonomous long-running sessions
- Want pricing users discuss favorably
Choose Aider if
- Work in a codebase with thousands of files
- Want manual control over which files change
- Prefer open source with local data handling
Where each one wins and loses
Why developers switched
Picked OpenAI Codex
No explicit switches recorded.
Picked Aider
Summaries are paraphrased. Each links to the original comment.