Gemini wrote the wrong code
- In many cases, the initial code test is more complicated than Gemini will always write it wrong. This is the same as GPT o1. I advise everyone to wait more or use Claude 3.5 Sonnet, the best coding model currently.
- Sorry for the grammar I don’t have very good English