GPT-5.5 vs Claude Opus 4.8: what 20 developers chose.
370 opinions. 20 developers named a direct preference: GPT-5.5 12–8. Approval is 69% vs 49%.
GPT-5.5 wins the direct preference 12–8, but both tools score best on the same topic. Use the decision cards below to choose on workflow.
Choose GPT-5.5 if you…
- Want the stronger overall and 90-day approval record
- Prioritize speed and lower cost
- Want the larger firsthand evidence base
Choose Claude Opus 4.8 if you…
- Prefer code that is easier to review
- Need the model to grasp intent on Figma or data work
- Can tolerate slower responses, higher cost, and tighter limits
GPT-5.5 and Claude Opus 4.8: 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 | GPT-5.5 | Claude Opus 4.8 | Mentions |
|---|---|---|---|
| Code quality | +60 | +13 | 114 / 62 |
| Reliability | +6 | -48 | 32 / 27 |
| Cost & pricing | +11 | -25 | 18 / 16 |
| Autonomy | +20 | -20 | 15 / 5 |
| Speed | +86 | -60 | 14 / 5 |
| Usage limits | +20 | -20 | 5 / 5 |
| UX | -33 | -50 | 3 / 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 GPT-5.5 12 developers
Opus 4.8 was extremely slow and went in circles, while GPT-5.5 finished comparable work faster.
Rates GPT-5.5 the better programmer for thoroughness, though prefers Opus 4.8's more human-friendly code and comments.
Ran both models on a week of workplace tasks and found GPT-5.5 clearly better, so they migrated.
GPT-5.5 solves problems better than Opus 4.8, though its generated code is over-engineered and uses shortcuts.
Picked Claude Opus 4.8 8 developers
Opus 4.8 is much better at coding and caught a sophisticated bug in a large codebase that GPT-5.5 missed.
Keeps Opus 4.8 for Figma front-end work and data analysis because it better understands the intent behind questions.
Prefers reading Opus 4.8 code over GPT-5.5 output because it is easier to review.
>equally "frontier" models A key point I want to make is that the notion of "frontier" is somewhat fictive in the sense that a model which dominates all others on a given eval is not guaranteed to be number one on another eval, even if both evals are ostensibly for the same task. For example, the best publicly-available model (i.e. excluding Claude Mythos and Fable) on DeepSWE[0] is gpt-5.5-xhigh at 67%, which is soundly better than claude-opus-4.8-max at 59%. I would say an 8pp gap on a benchmark is quite large. But on FrontierCode[1], claude-opus-4.8-xhigh is the best, at a score of 13.4% co
The numbers.
| Metric | GPT-5.5 | Claude Opus 4.8 |
|---|---|---|
| Approval | 69% | 49% |
| Last 90 days | 67% | 57% |
| Opinions | 236 | 134 |
| Developers | 221 | 128 |
| First-hand opinions | 203 | 117 |
| Direct picks | 12 | 8 |
| Type | Model | Model |
| Made by | — | — |
Related matchups.
GPT-5.5 vs Claude Opus 4.7
12–3 direct picks, and a clear difference in what developers optimise for.
GPT-5.5 vs Claude Fable 5
8–5 direct picks, and a clear difference in what developers optimise for.
Claude Fable 5 vs Claude Opus 4.8
9–3 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.