–

wapi · No library run 3

40.1Library score
Pass rate72.0%
Simplicity65.3
Problems $$4.94

Implementers get no library and write everything from scratch.

ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE78.4%63.7$0.83
45.7
GLM 5.3 Flash · mini-SWE72.9%69.1$2.49
41.4
GPT-5.6 Luna · Codex64.7%63.1$1.62
33.2
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean
agent-runtime-protocol2.146.70.216.3reservation-receipts71.30.00.023.8page-publish-links1.66.625.411.2catalog-api39.928.723.030.5portal-routing60.252.245.052.5index-gateway55.558.960.458.3chain-index54.658.764.459.2chain-archive71.265.418.151.6project-routes49.741.137.042.6callback-registry50.955.458.554.9