What 50,000 Runs of a 5-Line Eval Taught Us

https://code.visualstudio.com/blogs/2026/06/19/what-50000-runs-taught-us

4 points · 1 comments · view on lemmy.world

1 Comments

Lodra@programming.dev · 2 pts · 53d

Was anyone else annoyed that they don’t say which models they tested with? It’s all so secretive. Or did I miss that info somewhere?