Hacker Newsnew | past | comments | ask | show | jobs | submit | rusticpenn's commentslogin

It uses hidden models, mostly Claude and open ai models. They try to check the prompt and route to the model that is capable of it, thus saving money.

I think this is it ^^^ Routing and optimizations :)

Models are bad at chess. I am using a middleman to help models play chess and experimenting. https://abhay-ai.github.io/R_Daneel_AI/


I dislike crypto as much as the next guy but, the normal currencies are not very different.


Agreed apart from the fact that most currencies don't need to be converted to another currency first to determine which goods I can exchange for them. I agree that most currencies have no inherent value, only an agreed upon value.


That is a chicken and egg problem. Look at argentina pre Miley, when the dollar/peso exchange rate was fixed and you couldn't freely exchange. People would still prefer the dollar over the peso, even though you couldn't legally agree on dollar prices in contracts such as rentals. So people were using the dollar to save in, and exchange for peso when they needed to exchange goods.

My point being, you can have two competing currencies and exchange between them. In Argentinas case, the currency competition was healthy as politicians couldn't conceal their bad financial policies due to the hyper inflated exchange rate for pesos.


It depends on how much you want the LLM to do. I personally work on function level and can easily verify if it works with a look and few tests.


The whole field of metaheuristic algorithms rests on a similar idea. a lot of stupid "agents" finding a good solution. by metaheuristics i mean genetic algorithms, PSO, ACO etc.


They signed a contract with IBM to manufacture their chips at 7nm (and reneged).


Yeah and IBM had to move their designs in the last minute from GF to Samsung. I have heard that the Samsung process was much better and the tech transfer was easier than expected.


Using Neural nets, Deep Learning in pharma is not something new. Naturally LLMs are different. You should look into personalized medicine.


Thats WatsonX for you now.


I use Ecosia and it feels much better to me.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: