| GPT-5.6(openai.com) |
| 1342 points by logickkk1 18 hours ago | 934 comments |
| tl;dr: Summary not available |
| HN Discussion:- ↑Highlights useful semantic tips and benchmark achievements from the model's documentation
- ↓Skeptical of benchmark comparisons being cherry-picked or exclusions being convenient
- ↓Confused or frustrated by the naming convention (Sol/Terra/Luna) and model selection
- ~Independent testing shows model is comparable but not superior to competitors like Sonnet 5
- ↑Appreciates the focus on token efficiency alongside intelligence improvements
|
|