Claude Code vs OpenAI Codex: what 243 developers chose.
5,039 opinions. 243 developers named a direct preference: OpenAI Codex 181–62. Approval is 54% vs 69%.
If you want autonomy, choose Claude Code. If you want code quality, choose OpenAI Codex. The 181–62 direct split is not a landslide; it is a difference in working style.
Choose Claude Code if you…
- Want a mature subagent and plugin ecosystem
- Value Claude Code's terminal interface and permission flow
- Can absorb tighter usage limits for stronger autonomous runs
Choose OpenAI Codex if you…
- Need more useful work per dollar and fewer quota interruptions
- Want stronger code quality and review results
- Prefer a more stable harness with clearer progress display
Autonomy for Claude Code. Code quality for OpenAI Codex.
Net sentiment per topic: praise minus complaints, as a share of mentions. Positive means more praise than complaints. Hover a dot for the raw counts.
Net sentiment = praise minus complaints, as a share of each tool's mentions. The dot lanes are only for readability; position is the score.
View data table
| Topic | Claude Code | OpenAI Codex | Mentions |
|---|---|---|---|
| Autonomy | +52 | +57 | 368 / 215 |
| Code quality | +12 | +49 | 429 / 459 |
| UX | -3 | +21 | 437 / 217 |
| Reliability | -47 | -6 | 403 / 239 |
| Cost & pricing | -18 | +53 | 313 / 228 |
| Usage limits | -44 | +36 | 228 / 192 |
| Speed | +48 | +49 | 169 / 129 |
| Customization | +2 | +29 | 178 / 51 |
The same decision, from both sides.
Paraphrased from comments where a developer named a preference or described a switch. Each links to the original.
Picked Claude Code 62 developers
Claude Code has the better UX, while Codex chases mass-market design and suffers for it.
Reports Codex filled about 12 GB with old versions, and rates it worse than Claude Code on every metric.
Codex quality and reliability have declined, and its coding falls short of Claude Code with Opus.
Codex cloud agents force you onto your own machine, while Claude Code works better as a cloud dev environment.
Picked OpenAI Codex 181 developers
Opus drains a weekly quota in a day or two while Codex keeps running, and finds more flaws in Claude's commits.
Switched from Claude to Codex because the Claude Code harness feels unstable and frustrating.
> They don’t really. Most normies which are the vast majority of people just use ChatGPT. All tech folk just use Claude. It was the case a year ago, but it's less clear cut nowadays, I've seen non-tech people using claude on their phone in the train, and many tech people seem to have switched from Claude Code + Opus to Codex + Sol over the past few month.
Picks Codex for more generous usage limits and resets, plus unmetered ChatGPT access on the 20x plan.
The numbers.
| Metric | Claude Code | OpenAI Codex |
|---|---|---|
| Approval | 54% | 69% |
| Last 90 days | 51% | 63% |
| Opinions | 3,026 | 2,013 |
| Developers | 2,204 | 1,386 |
| First-hand opinions | 2,397 | 1,708 |
| Direct picks | 62 | 181 |
| Type | Terminal agent | Terminal agent |
| Made by | Anthropic | OpenAI |
| Website | Visit ↗ | Visit ↗ |
Related matchups.
Claude Code vs OpenCode
73–42 direct picks, and a clear difference in what developers optimise for.
Claude Code vs Cursor
61–42 direct picks, and a clear difference in what developers optimise for.
Claude Code vs Pi
54–6 direct picks, and a clear difference in what developers optimise for.
How to read this comparison.
Approval is the positive share of opinions that took a side, with the rank based on the lower bound of a 90% confidence interval. Direct picks count comments where a developer explicitly preferred one tool or described switching. Small samples are noisier; check the opinion counts.
Get the monthly report
Which AI coding tools gained or lost developer approval on Hacker News, and why people switched. One email a month.