Y
Hacker News
new
|
ask
|
show
|
jobs
by
zlatkov
226 days ago
I haven’t come across any research showing that a specific LLM consistently outperforms others for this. It generally works best with strong reasoning models that produce consistent outputs.