OpenAI Codex vs Gemini CLI: what 12 developers chose.
2,224 opinions. 12 developers named a direct preference: OpenAI Codex 12–0. Approval is 69% vs 29%.
If you want code quality, choose OpenAI Codex. If you want autonomy, choose Gemini CLI. The 12–0 direct split is not a landslide; it is a difference in working style.
Choose OpenAI Codex if you…
- Want stronger code-quality feedback
- Want more positive autonomy and usage-limit feedback
- Want a larger HN evidence base
Choose Gemini CLI if you…
- Want an open-source terminal agent
- Work inside an existing IDE or notes vault
- Need a fallback when Codex limits are reached
Code quality for OpenAI Codex. Autonomy for Gemini CLI.
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 | OpenAI Codex | Gemini CLI | Mentions |
|---|---|---|---|
| Code quality | +49 | -48 | 459 / 23 |
| Autonomy | +57 | -10 | 215 / 20 |
| Reliability | -6 | -77 | 239 / 62 |
| UX | +21 | -60 | 217 / 35 |
| Cost & pricing | +53 | -22 | 228 / 18 |
| Usage limits | +36 | -85 | 192 / 13 |
| Speed | +49 | -33 | 129 / 6 |
| Customization | +29 | -50 | 51 / 4 |
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 OpenAI Codex 12 developers
Gemini CLI helps check specifications for drift, but is weaker than Codex for general coding work.
Calls Gemini CLI's coding performance poor and says it barely gets used.
Gemini CLI handles agentic tasks badly and fails at tool calls.
Gemini CLI gets stuck in loops and does not match the Codex experience.
Picked Gemini CLI 0 developers
No explicit switches recorded.
The numbers.
| Metric | OpenAI Codex | Gemini CLI |
|---|---|---|
| Approval | 69% | 29% |
| Last 90 days | 63% | 25% |
| Opinions | 2,013 | 211 |
| Developers | 1,386 | 175 |
| First-hand opinions | 1,708 | 163 |
| Direct picks | 12 | 0 |
| Type | Terminal agent | Terminal agent |
| Made by | OpenAI | |
| Website | Visit ↗ | Visit ↗ |
Related matchups.
Claude Code vs OpenAI Codex
181–62 direct picks, and a clear difference in what developers optimise for.
Claude Code vs Gemini CLI
32–4 direct picks, and a clear difference in what developers optimise for.
OpenAI Codex vs Cursor
21–7 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.