OpenAI Codex vs Cursor: what developers actually say
Based on 2,779 opinions on Hacker News from developers who discussed these two, most of them first-hand users.
Codex comes out ahead with 69% approval versus 55%, 63% versus 53% over 90 days, 2,013 opinions versus 766, and 21 explicit picks versus 7. Its aspect counts favor cost/pricing (174 positive, 54 negative), usage limits (131/61), and code quality (343/116), while Cursor leads on speed (44/8) and slightly on UX balance (101/59 versus 131/86). Cursor's reliability is 14 positive and 58 negative, and its cost score is split at 58/60. It is the better pick for IDE-first work with fast autocomplete, while Codex suits terminal-first work and users who care about cost and limits.
When developers explicitly said they prefer one or switched: 21 picked OpenAI Codex, 7 picked Cursor.
Choose OpenAI Codex if
- Want stronger cost and usage-limit scores
- Prefer a terminal-first agent
- Want more first-hand reports and higher approval
Choose Cursor if
- Want an IDE-first workflow
- Prioritize speed and autocomplete
- Prefer editor integration over terminal work
Where each one wins and loses
Why developers switched
Picked OpenAI Codex
Picked Cursor
Summaries are paraphrased. Each links to the original comment.