Gonna play around with it for my publishing use cases. It is already straightforward to use to compare output from different models to the same prompt

→ View original post on X — @testingcatalog