This is grok 3, so not a debut
Debut in the sense that it’s something good enough that it’s getting mainstream attention.
Unfortunately LLMs are shifting compute time to test time instead of train time. I don't really like this and frankly it shows a stalling of the architectures, data sets, etc...
Naively, I feel to be useful, the goal of LLMs should be to more power efficient. So that eventually all devices can be smarter.
Power efficiency can be gained through less time-time, or more "intelligence" or some combination of the two. I'm not convinced these SOTA models are doing much more than increasing test-time.