As someone who doesn't spend a huge amount of time thinking about art, this piece soothed me more than I could have expected. Thank You

yeah but not super fast like flash or grok fast

I’m sure they do something like that. I’ve noticed azure has way faster gpt 4.1 than OpenAI

I'm pretty sure xAI exclusively uses Nvidia H100s for Grok inference but I could be wrong. I agree that I don't see why TPUs would necessarily explain latency.

I just want a deepseek moment for an open weights model fast enough to use in my app, I hate paying the big guys.