–
uglypie · No library run 3
31.6Library score
Pass rate86.4%
Simplicity40.8
Problems $$2.70
Implementers get no library and write everything from scratch.
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE85.9%39.5$0.74
GLM 5.3 Flash · mini-SWE81.3%43.0$1.25
GPT-5.6 Luna · Codex92.6%40.0$0.71
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean
portal-ingest54.246.238.146.2kml-schema-import29.832.530.430.9news-crawl24.815.70.013.5access-tree18.928.027.924.9page-cleaner56.646.847.250.2exhibit-route19.920.523.521.3hearing-replay37.136.541.238.3festival-revisions31.534.932.232.9page-notes0.00.025.08.3genome-catalog45.259.044.649.6article-rewrite32.728.432.431.2