The business bottom line depends on tokens, shareholders want to see exactly that.

With how competitive the LLM field is, it would surprise me greatly if any of these players were doing anything other than trying to make the best possible product. I certainly do not believe they are intentionally training the models to use more tokens unnecessarily.

▲

owebmaster an hour ago | parent | next [-]

There is a theory that the verbosity and comments help getting better results with the current benchmarks. So the models are theoretically getting better but in practice they are getting worse.

▲

hmokiguess 4 hours ago | parent | prev | next [-]

Are you implying they can’t do both?

	▲	daishi55 4 hours ago \| parent [-]
		Yes, those things are mutually exclusive.

▲

cyanydeez 4 hours ago | parent | prev | next [-]

Theyre training the system to minimize compute,so most likely theyre dynamically downgrading quants in the first few turns hoping to find the cheapest model to run. The side effect may be excessive token gen

▲

daishi55 4 hours ago | parent [-]

Sure. That’s possible. But that’s not what I was talking about.

	▲	cyanydeez 7 minutes ago \| parent [-]
		Tomato, technical tomato

▲

LaGrange 3 hours ago | parent | prev [-]

You're joking, right? That's an incredibly outmoded view of capitalist "competition."