Claude Opus 4.5 vs GPT-5.2: what developers actually say
Based on 404 opinions on Hacker News from developers who discussed these two, most of them first-hand users.
Claude Opus 4.5 comes out ahead on the evidence: 322 opinions from 259 authors against 82 from 77, approval of 78 against 68, and a 90-day approval of 90, while GPT-5.2 has no 90-day figure. Its code-quality aspect runs 120 positive to 28 negative, about 81% positive, against 70% for GPT-5.2, and it draws more praise for speed, autonomy and large codebases. GPT-5.2 wins the explicit head-to-head 7 to 3, where users cite lower cost, more generous usage limits and steadier long-running agents. Choose GPT-5.2 when budget or quota is the constraint and work runs unattended; choose Opus 4.5 for interactive coding and stronger code output.
When developers explicitly said they prefer one or switched: 3 picked Claude Opus 4.5, 7 picked GPT-5.2.
Choose Claude Opus 4.5 if
- Want stronger code quality and faster output
- Steer the model interactively on complex codebases
- Want the model with more discussion and higher approval
Choose GPT-5.2 if
- Need more usage per dollar and a generous plan
- Run long unattended agent tasks that must follow instructions
- Want a reviewer for code written by Claude
Where each one wins and loses
Why developers switched
Picked Claude Opus 4.5
Picked GPT-5.2
Summaries are paraphrased. Each links to the original comment.
Side by side
| Claude Opus 4.5 | GPT-5.2 | |
|---|---|---|
| Approval | 78% | 68% |
| Last 90 days | 90% | – |
| Opinions | 322 | 82 |
| First-hand users | 254 | 73 |