P
weft · Production library run 1
30.6Library score
Pass rate66.0%
Simplicity53.7
Problems $$17.63
Implementers use the real production library (chumsky).
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE46.5%49.5$1.96
GLM 5.3 Flash · mini-SWE71.6%54.9$12.18
GPT-5.6 Luna · Codex79.9%56.6$3.50
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean
Click a problem to see how each solution's score is built.