Add Handbook.md benchmark tasks

2026-06-24 12:44:34 -07:00
commit 72f9a65917
1800 changed files with 323589 additions and 0 deletions
@@ -0,0 +1,4 @@
+OPENROUTER_API_KEY=your-openrouter-key-here
+# Use anthropic API directly, openrouter for other models. 
+# reasoning settings were not supported by openrouter at eval time. 
+ANTHROPIC_API_KEY=your-anthropic-key-here