Show HN: A/B test Gemini, ChatGPT and Claude (news.ycombinator.com)

2 points by eclair99 ↗ HN
I built this app for macOS to A/B test LLMs side-by-side. Also, I feel that having your prompt answered by two LLMs is a quick and easy way to confirm you are not falling for hallucinated information.

3 comments

[ 3.0 ms ] story [ 19.0 ms ] thread
(comment deleted)
Any insights into which is better from the using the app?
Thanks for checking out the tool!

I personally feel Claude3 outperforms Gemini and ChatGPT. But lack of web-browsing is a slight dis-adv of claude.