GPT-5.6 Sol vs Claude Opus 5: what developers actually say
Based on 409 opinions on Hacker News from developers who discussed these two, most of them first-hand users.
GPT-5.6 Sol comes out ahead on the weight of the data. It has 247 opinions from 206 authors and 67% approval, compared with 162 opinions and 33% for Claude Opus 5, and 7 commenters picked it over Opus 5 against 3 the other way. Its code quality mentions run 73 positive to 26 negative, and it leads on cost (22 to 4), speed (14 to 2) and UX (6 to 1). Claude Opus 5 is the better second model for review and planning work, for a lower output token price, and when you can give tight constraints and check the output.
When developers explicitly said they prefer one or switched: 7 picked GPT-5.6 Sol, 3 picked Claude Opus 5.
Choose GPT-5.6 Sol if
- You want the higher overall approval and the larger body of reported use
- Code quality, speed and UX mentions carry the most weight
- You want the model most commenters in this sample switched to
Choose Claude Opus 5 if
- You need a second model for review and planning
- Per-token output price matters
- You can constrain it tightly and verify its work
Where each one wins and loses
Why developers switched
Picked GPT-5.6 Sol
Picked Claude Opus 5
Summaries are paraphrased. Each links to the original comment.
Side by side
| GPT-5.6 Sol | Claude Opus 5 | |
|---|---|---|
| Approval | 67% | 33% |
| Last 90 days | 66% | 30% |
| Opinions | 247 | 162 |
| First-hand users | 210 | 136 |