767 karma · joined November 23, 2022
The tone of the grandparent comment made it sound like the author is a child, my bad.
But that was 2 days ago (and the site was struggling then too), I wonder what it is this time.
If designed well - by suspending generation in memory and inserting a <function_result>, without restarting generation and fetching cache from disk - the round trip/tool call is better (costs the equivalent of 50 tokens for waiting + function_result tokens).
Granted, you could also just import * from './low-level.wat' (or .c, and compile it automatically to WASM)
Maybe say which country specifically you mean when you say "dual citizenship" ;)
I enjoy "bloated" languages. Many languages are bloated nowadays, but the community agrees what set of features to use, what to avoid. Still, those rare features can be useful for stuff like making DSLs through clever use of the language.
It's much worse to have a minimal language that takes years to add new features, like how Go took over a decade to add generics.
I wish this was extended to things like: you could give the model an API endpoint that it can call to execute JS code, and the only requirement is that your API has to respond within 5 seconds (maybe less actually).
I wonder if this is what OpenAI is planning to do in the upcoming API update to support tools in o3.