Claude Opus 4.6 vs GPT-5.4: what 14 developers chose.
401 opinions. 14 developers named a direct preference: GPT-5.4 10–4. Approval is 63% vs 72%.
GPT-5.4 wins the direct preference 10–4, but both tools score best on the same topic. Use the decision cards below to choose on workflow.
Choose Claude Opus 4.6 if you…
- Want stronger instruction-following and analysis
- Value formatted output and UX
- Work in Qt C++/QML or want the model to check existing code first
Choose GPT-5.4 if you…
- Want higher approval and the 10 to 4 head-to-head edge
- Care about reliability, cost, and large-codebase work
- Need a coding model that follows plans to completion
Claude Opus 4.6 and GPT-5.4: where each one wins.
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 Opus 4.6 | GPT-5.4 | Mentions |
|---|---|---|---|
| Code quality | +29 | +45 | 121 / 51 |
| Reliability | -34 | +18 | 58 / 17 |
| Autonomy | +45 | +43 | 29 / 7 |
| Cost & pricing | 0 | +71 | 14 / 14 |
| Large codebases | 0 | +100 | 8 / 6 |
| UX | +60 | -33 | 5 / 3 |
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 Opus 4.6 4 developers
The commenter found Opus 4.6 stronger outside straight coding, especially for analysis, planning, thoroughness, and formatted output.
Said Opus 4.6 followed instructions better and gave more consistent results, while GPT-5.4 often forgot instructions and went on tangents.
A Qt C++ and QML developer found Opus 4.6 thinking much better than GPT-5.4 and the other models tested.
Said GPT-5.4 made unrequested refactors and introduced a bug, while Opus 4.6 recognized the feature already existed and proposed tests and docs.
Picked GPT-5.4 10 developers
Said Opus 4.6 followed trained habits over user intent and struggled with old papers, making GPT-5.4 the better fit.
Said GPT-5.4 xhigh beat Opus 4.6 in repeated head-to-head tests and was a stronger implementer, though Opus worked as a reviewer.
Said GPT-5.4 found deeper problems in a refactor than Opus 4.6, though the commenter still preferred Opus for writing code.
Said GPT-5.4 used about half the tokens and followed plans to completion, while Opus 4.6 took shortcuts.
The numbers.
| Metric | Claude Opus 4.6 | GPT-5.4 |
|---|---|---|
| Approval | 63% | 72% |
| Last 90 days | 63% | – |
| Opinions | 287 | 114 |
| Developers | 257 | 105 |
| First-hand opinions | 242 | 101 |
| Direct picks | 4 | 10 |
| Type | Model | Model |
| Made by | — | — |
Related matchups.
Claude Opus 4.6 vs Claude Opus 4.7
12–2 direct picks, and a clear difference in what developers optimise for.
Claude Opus 4.5 vs Claude Opus 4.6
7–4 direct picks, and a clear difference in what developers optimise for.
Claude Opus 4.6 vs GPT-5.3 Codex
5–2 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.