Guanlan Dai @guanlan · 02 Sep 2026
A year ago the question was which model. Now it's which harness. Pi, Exo, Claude Code, Codex, DeepSeek Harness and 4 others. Same model, same tasks, same runtime. 360 runs, 2 billion tokens. Pass rates: 50% to 67%. Cost per pass: $1.05 to $18.34. Introducing FrontierHarness https://t.co/4QCzPyCmYc
285 607Views
1 977Likes
209Reposts
198Replies
75Quotes
1 650Bookmarks
Is that a lot?
15.8×vs this author's median18 132 views is typical
80Percentile for this authorof 5 recent posts
39.9×vs under 10K median7 156 views is typical
48.3× audienceReachviews ÷ followers
0.86%Engagement rateof viewers reacted