This new tool will allow you to run user and system prompts on 4 models at once and compare the output, latency and cost for each. You will be able to choose from over 300 models to try! It’s sort of like the Langfuse playground but you don’t need an account and can just start right away.

I’ve also added the ability to favor responses and will later aggregate up and display user preference data inspired by LMArena.

Check it out here!

Looking forward to your feedback! Feel free to email me here or use the widget I have on the site!

Best,

Alex