LLMs Are Too Big. My Log Router Doesn't Need to Sing.
This story is from 2026-09-22. It is preserved in the archive; the latest stories are on the live feed.
I think our language models are too big. Which is a little bit spicy to say, because I don't actually mean the parameter count. I mean that we've gotten comfortable asking an incredibly general system to do a very specific job, and then doing a lot of work to keep it focused on that job. Working on…
Read the full story at DEV Community — AI ↗
Timeline · 1 report
- 2026-09-22 18:01 · DEV Community — AI
LLMs Are Too Big. My Log Router Doesn't Need to Sing.