Fable 5.1

42.6± 2.5 (95% CI)

Claude Code · pass rate 85.8% · simplicity 58.2 · $14.39 per library

Runs

Each cell is one full run: the model designs the library, then 3 implementers solve every problem with it. Click a cell to open that run.

TaskRun 1Run 2Run 3Mean ± std

arazu 39.7 39.9 15.231.6 ± 14.2

clirs 58.6 58.4 66.361.1 ± 4.5

heretical 46.5 48.4 44.046.3 ± 2.2

npr 51.5 50.7 52.751.6 ± 1.0

plaid 46.3 47.5 48.247.3 ± 1.0

pyda 57.6 57.1 37.250.6 ± 11.6

canon 49.4 58.6 29.245.7 ± 15.0

roadkill 12.8 11.4 11.912.0 ± 0.8

rubberband 76.2 73.0 74.174.4 ± 1.6

weft 37.8 33.8 37.136.2 ± 2.2

sapi 19.4 41.1 39.433.3 ± 12.1

simu 44.1 40.5 40.941.8 ± 1.9

uglypie 28.6 48.2 31.035.9 ± 10.7

umami 49.5 48.2 47.048.3 ± 1.3

wapi 20.8 26.1 23.123.3 ± 2.7