GPT-5.5 vs Claude Opus 4.7: what developers actually say
Based on 381 opinions on Hacker News from developers who discussed these two, most of them first-hand users.
GPT-5.5 comes out ahead overall: 69% approval from 236 opinions versus 42% from 145, and 12 explicit preferences to 3. Its code-quality counts are 91 positive and 23 negative, against 23 and 28 for Opus 4.7, and it also leads on speed (13 positive, 1 negative versus 5 and 2), reliability (17 and 15 versus 8 and 26), cost (10 and 8 versus 3 and 9), and usage limits (3 and 2 versus 1 and 7). Opus 4.7 is the better pick when planning and architecture dominate, based on one user who valued its pushback on bad ideas; its 90-day approval is missing and its monthly opinion volume falls to 3 by August 2026. Choose Opus 4.7 only when that planning strength and its style outweigh weaker quality, reliability, cost, and limit sentiment.
When developers explicitly said they prefer one or switched: 12 picked GPT-5.5, 3 picked Claude Opus 4.7.
Choose GPT-5.5 if
- Want stronger code-quality sentiment and higher approval
- Care about speed and reliability
- Want more evidence from a larger comment base
Choose Claude Opus 4.7 if
- Put planning and architecture first
- Prefer a model that pushes back on bad ideas
- Can absorb more cost and limit complaints
Where each one wins and loses
Why developers switched
Picked GPT-5.5
Picked Claude Opus 4.7
Summaries are paraphrased. Each links to the original comment.
Side by side
| GPT-5.5 | Claude Opus 4.7 | |
|---|---|---|
| Approval | 69% | 42% |
| Last 90 days | 67% | – |
| Opinions | 236 | 145 |
| First-hand users | 203 | 122 |