Skip to main content

Claude 3.5 is worse than Gemini 1.5 pro

Feature Request

I thought claude was meant to be better than gemini since it generally is in the llm benchmarks but claudes code explanation is really bad and doesn’t sound human at all meanwhile gemini 1.5 pro is and provides very in depth explanation

Posted by Nova ·
4 comments

Comments4

  • NASI

    •

    Oct 31, 2024

    我想我也遇到了同样的问题，在使用中我发现无论是gpt4或是claude，给出的答案似乎都不如gemeni 1.5pro更让人满意，也许这跟它的家庭google有关？我不知道其他平台的claude和gpt4是否更“聪明”一些？个人的主观评价，不一定符合真实模型能力，仅供参考

  • Nova

    •

    Sep 2, 2024

    I bought it thinking it would be an upgrade from perplexity more coding focused. But I have perplexity pro and yet the questions devv.ai claude 3.5 sonnet gives is somehow worse than perplexity’s claude 3.5 sonnet in terms of beginner friendly answers