Let's rate AI (LLMs) services based on real-world experience, not just academic papers and benchmarks.

Hi, folks! As an AI chatbot user, I really enjoy using it to assist me with writing, coding, research, etc. I've been using GPT-3.5 Turbo since its release and occasionally use GPT-4. I'm not an AI "fanatic" nor interested in the academic papers and benchmarks. In the community, some people say X is better than Y, while some people say the opposite, which makes me confused. Finally, I exhausted my enthusiasm and stuck with my GPT-3.5 combo. So, I built this website to help ordinary chatbot users rate AI models based on their real-world experiences, and discover the best option for their requirements. Edit 1: I know lmsys :D. It has a nice blind testing. But I think it's still a bit academic, and it's not straightforward for non-tech users who just want to rate for certain aspects. Edit 2: CAPTCHA is currently used to minimize the possibility of cheating. And to ensure timeliness, only ratings from the last 90 days are counted.

congrats on the launch, it's a practical approach to understanding ai tools. with diverse user feedback in play, how do you plan on managing the accuracy and helpfulness of the reviews to aid users in making the best choice for their needs?

Hi @mashy . That's a fair question. CAPTCHA is currently used to minimize the possibility of cheating. And to ensure timeliness, only ratings from the last 90 days are counted.

You Rate AI

Rating AI based on real-world experience

Rating AI based on real-world experience