Richard A Aragon

TuringsSolutions

AI & ML interests

None yet

Articles

Organizations

Posts 6

view post
Post
1272
SNN Image Diffusion V2

Billionaires have been made for less than this. This is only one of the things it can it do. It can do API calls, function calls, optimize poker and blackjack odds, anything that is an optimization problem. It costs fractions of a penny and requires fractions of the compute of an LLM model. It can even communicate two ways with an LLM model.
view post
Post
442
I can do Time Series Predictions with Swarm Algorithms! When all you know how to use is a hammer, everything looks like a nail. An LLM model is a hammer. It is not a deity. It has computational and mathematical limitations. Very big ones. Swarm Algorithms do not have this same problem. They are like a screwdriver. The screwdriver is not better than the hammer, both are useful. Why are LLM models bad at things like Time Series Predictions and Function Calls? Because those are jobs better fit for a screwdriver as opposed to a hammer.