Y
Hacker News
new
|
ask
|
show
|
jobs
sourcecodeplz
8 hours ago
[
-
]
still quite verbose at 140m output tokens, but this is on max thinking. high should do better.
Please enable JavaScript to continue using this application.