> What is needed the most right now is something similar to Bonsai 27B, with a modest memory footpint, but faster and more capable

Yeah, that'd be neat, but that's not what this announcement is about at all:

> With a massive 2.4T parameters

▲

docheinestages a day ago | parent | next [-]

True. It was more of an open letter, with hopes that the Qwen team sees the comments in this thread.

▲

cyanydeez a day ago | parent | prev [-]

dont we all deem the ability to improve large models as the defacto capability to produce small ones?

▲

embedding-shape a day ago | parent [-]

I don't think so, they have different constraints and require different optimizations, being able to produce one of them doesn't mean you'll automagically be good at the other.

	▲	cyanydeez a day ago \| parent [-]
		but that's denying the singularity boostrap theory. Which I don't agree with, but if you can't harness a large model to make a small model, then we're going to have problems brining about the singularity. If you do think there's some magical singularity, how do you comport?