P
wapi · Production library run 2
32.8Library score
Pass rate57.4%
Simplicity75.8
Problems $$20.96
Implementers use the real production library (servant).
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE42.3%73.9$1.65
GLM 5.3 Flash · mini-SWE60.0%74.9$15.26
GPT-5.6 Luna · Codex70.2%78.4$4.04
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean
agent-runtime-protocol2.80.039.414.1reservation-receipts0.035.00.812.0page-publish-links3.88.64.25.6catalog-api32.615.927.125.2portal-routing92.627.254.258.0index-gateway86.454.686.475.8chain-index0.052.761.238.0chain-archive58.735.430.441.5project-routes0.00.058.619.5callback-registry3.237.174.138.1