Gemini is worse than GPT-3.5 in my experience

I use GPT-3.5 for simple productivity and finding information on websites. I made the switch to Gemini after my GPT-3.5 token was invalidated. I've been using Gemini for over a year and it is awful. <br><br>I no longer have my token so I can't test it, but I can confirm the Gemini worse than GPT-3.5 in these three ways, in my experience:<br><br>1. **Accuracy**: Gemini is significantly less accurate than GPT-3.5. GPT-3.5 was worse than GPT-4 in my experience but generally speaking, the information produced by GPT-3.5 was accurate and correct. Gemini is so inaccurate that it makes me question whether or not it simply fabricates information. It regularly looks up the same info on the same websites that GPT-3.5 was able to do correctly, but provides the wrong information. It seems to do this more often than GPT-3.5. I have even tested it by feeding it information already on a page and then asking it to look up that information on the page. The information it provides back is regularly fabricated and inaccurate.<br><br>2. **GPT-4 like responses to simple questions**: Gemini is so bad that it gives me GPT-4 like responses to simple questions. Simple questions like 'how much does the atmel microcontroller cost' will give me the usual GPT-4 response of "The Atmel microcontroller is a series of microcontrollers that are manufactured by Microchip Technologies, Inc. (formerly Atmel Corporation). There are several different types of Atmel microcontrollers, each with its own unique characteristics and features. It is not possible to provide a single price for an Atmel microcontroller without more specific information. Could you please provide more details or clarify your question?"<br><br>3. **Failure to locate information**: Gemini fails to locate information that, in my experience, GPT-3.5 could locate with ease. I've tested this on multiple websites that do not require logins. I've even tested it on pages with very few pieces of information. It seems to me that Gemini is either not reading my prompts correctly or is failing to locate information correctly.<br><br>I do have to say that Gemini is a lot faster than GPT-3.5. It will often respond to my prompts almost instantly. So I suppose you get what you pay for.