请问HN：OpenAI模型与Gemini 2.5 Pro在编码和软件工程方面的比较
根据您的经验，在作为助手回答软件工程/软件系统相关问题以及进行长时间复杂推理方面，OpenAI的模型和Gemini 2.5 Pro这两者中哪个更好？我在考虑是否有必要为ChatGPT付费，还是使用Gemini 2.5 Pro的免费版本或付费版本更合适。我感觉大多数Hacker News的用户更倾向于后者，但在livebench中，我认为OpenAI在编码方面超过了Gemini。
查看原文
In your experience, which of the two models (all of OpenAI vs Gemini 2.5 Pro) are better for having as assistants to ask SWE/software systems related questions and doing long and complex reasoning?<p>I'm debating whether there's any point in paying for ChatGPT vs. paying (or even using the free version) of Gemini 2.5 Pro.<p>I have the feeling that most HNers prefer the latter, however in livebench I think OpenAI surpasses Gemini for coding.