A wire transfer happened.

Depends on what you want to do. Some task specific models can be trained with a few ten or hundred thousand training examples so you can use a bigger model to produce synthetic training examples and then fine tune a smaller student model. I think that's the usual process. Whether you'd get acceptable performance this way depends, as mentioned, on what you're trying to do and what you'd consider acceptable.

once you are able to get the full probability distributions per token you can distill it on specific domains. distilling without that isn't generally a good idea unless you have invested millions in the requisite infrastructure.

which one of the other chinese frontier models is better?

DS4 Flash Q2/Q4 mixed quant fits on a DGX Spark (a $4000 device which is not particularly unheard of expense for Apple customers), and is indistinguishable for me from Opus for my personal daily use/assistant benchmarks[0].

[0]https://humanparadox.org/local-vs-frontier-benchmarks-for-my... - note here I tested Q8 but have found no difference at lower quant.

Apple is already doing this... they worked with Gemini to distill the model into a smaller one that fits on your phone. If you have iOS 27 Beta, you're already using this

Almost surely. Apple is extremely well positioned to take advantage of this over the next decade.