Hi guys,
Supper happy to submit our new app "LLM Bench" for Bubble AI challenge! This is an LLM comparison tool that solves a real problem I've faced as a founder.
Choosing between ChatGPT, Claude and others is tedious, slow and can lead to expensive business decisions - some LLMs cost 2x higher for same prompt.
We solved this with smart LLM split-testing tool - test 4 major LLM models simultaneously with same prompt. See response quality, speed, exact costs, and token usage - all in one comparison view.
Built by Bubble AI for dev's, founders and product teams who want to make data-driven AI decisions.
Live app URL: https://markomilojkovicaut-74973.bubbleapps.io/version-test/
Hopefully you guys will enjoy it and USE IT - we have a lot of cool features coming up. Feedback is very welcome!
Thank you, this was super fun build!