Benchmarks are favorable enough they're comparing to non-OpenAI models again. Interesting that tokens/second is similar to 5.4. Maybe there's some genuine innovation beyond bigger model better this time?
4,015 karma · joined September 30, 2014
Just a shockingly constrained service tier right now.
It's literally the first thing that came to mind when I saw your post and not having a convincing/satisfying answer in direct relation to that catastrophe doesn't bode well for getting people to trust your brand. The rest of your answer is essentially the absolute minimum I'd expect from a business like this, but not sufficiently convincing.
Not sure if this is more concerning for the test time compute paradigm or the underlying model itself.
Maybe I'm misunderstanding something though? I'm assuming 5.4 and 5.4 Thinking are the same underlying model and that's not just marketing.
I can't imagine a description more likely to give corporate workers the ick.