Skip to content
Better HN
Top
Best
Ask
Show
New
Jobs
Search
⌘K
0 points
sinuhe69
5d ago
0 comments
Save
Share
Recent incident with the Rio 3.5 model clearly shows that many coding models are specifically trained/fine tuned for the benchmarks.
0 comments
1 comments · 1 top-level
top
newest
oldest
RDTvlokip
4d ago
That's what I thought
j
/
k
navigate · click thread line to collapse