Guanlan Dai @guanlan · 02 Sep 2026
Here is the run that made us write this up. One of the hardest tasks, the pass rate (all model efforts) on DeepSWE is 38%, Pi and Claude Code both fixed it. Pi took 90 turns and $2.50. Claude Code took 381 turns and $64.36. Roughly 26x more for the same fix. https://t.co/VCX37wOxxm
22 118Views
145Likes
6Reposts
9Replies
4Quotes
57Bookmarks
Is that a lot?
1.22×vs this author's median18 132 views is typical
60Percentile for this authorof 5 recent posts
3.09×vs under 10K median7 156 views is typical
3.7× audienceReachviews ÷ followers
0.74%Engagement rateof viewers reacted