Hacker News new | ask | show | jobs
by nat0704 692 days ago
Seems you recently checked into 2024.

There are things called 'Open Weight' language models, a lot of which (those with between 8b to 34b parameters anyway) are considerably cheaper to run any of OpenAI's models.

1 comments

Moreover, you get surprisingly out-of-class (size-wise) performance if you fine-tune for your specific problem space. Even if you only train in a parameter-efficient way.