@GaryMarcus@lividprowess@zeratizerata I don't agree with the statement that pure LLMs have only marginally improved, but even still that is just unimportant to the fact that the usefulness of openai or anthropic serving their models have dramatically increased and nothing else really matters past that statement.
@lividprowess@zeratizerata@GaryMarcus this feels like unnecessary semantics at this point. the models have gotten better as well as the tools used to guide and serve them.
this is confused. it’s not LLMs that are good at math; it’s LLMs merged with (neuro)symbolic tools that are good at math.
amazing that people feel free to clutch their pearls without understanding even the basic technical details.
@deanwball The "ungovernability" (and openness) of Linux and the Internet is precisely what has made their success.
The same will be true of open weight AI foundation models.
the court-ordered psychiatrist measured my skull . he said there isnt medicine yet for what i have . he talked to ai on his computer for a while and then let me go. they gave me a bus ticket to San Francisco— said i’d be happy there
@iamgingertrash Based on my limited knowledge of your product:
are truffles going to run ~86 B models which all specialize in some specific task. Then there's going to be some method of communication where they act as a collective intelligence ?
Akin to humans connecting through the internet