Skip to content
Better HN
Top
Best
Ask
Show
New
Jobs
Search
⌘K
0 points
minimaxir
2mo ago
0 comments
Save
Share
These changes fixed
some
of the token issues, but the token bloat is an intrinsic problem to the model, and Anthropic's solution of defaulting to xhigh reasoning for Opus 4.7 just means you'll go through tokens faster anyways.
0 comments
1 comments · 1 top-level
top
newest
oldest
sgt
2mo ago
I'm worried anything less than xhigh is insufficient though. What do you do?
j
/
k
navigate · click thread line to collapse