How to Evaluate a Multimodal Model

Ben Popper: I'd say like a more tractable thing is just risking really hard to evaluate these models. Do you all work on anything that's multimodal or just LLM and just text based? Oh, we're all over multimodal. We have a stellar computer vision team. You can always find me on Twitter at Ben Popper. Reach us with questions for the podcast at Stack Overflow.