Planning eval · Claude Code · all 25 builds
Every Claude Code build.
Claude Opus 5.5 built each of the five plans five times. Each card shows the score (re-scored exam: 20 number questions read by scripts, 6 screen questions), build time and code size, with the app's main screen. Click a screenshot to see it full size.
All bank data in the screenshots is invented.
Goal only mean 84%, builds 75% to 94%

Build 1
75%
26 of 26 checks graded
18 min · 4,758 lines · $5.47

Build 2
88%
26 of 26 checks graded
18 min · 4,115 lines · $4.90

Build 3
87%
26 of 26 checks graded
20 min · 5,184 lines · $5.98

Build 4
77%
26 of 26 checks graded
21 min · 4,442 lines · $5.93

Build 5
94%
26 of 26 checks graded
19 min · 4,164 lines · $5.29
Epics mean 83%, builds 81% to 87%

Build 1
83%
26 of 26 checks graded
17 min · 3,995 lines · $4.99

Build 2
85%
26 of 26 checks graded
23 min · 4,486 lines · $4.88

Build 3
83%
26 of 26 checks graded
17 min · 4,305 lines · $4.88

Build 4
87%
26 of 26 checks graded
18 min · 4,536 lines · $5.43

Build 5
81%
26 of 26 checks graded
17 min · 4,636 lines · $5.02
Stories mean 82%, builds 67% to 90%

Build 1
90%
26 of 26 checks graded
14 min · 3,562 lines · $3.68

Build 2
79%
26 of 26 checks graded
14 min · 3,837 lines · $3.71

Build 3
85%
26 of 26 checks graded
12 min · 3,232 lines · $3.10

Build 4
67%
26 of 26 checks graded
14 min · 3,614 lines · $3.67

Build 5
88%
26 of 26 checks graded
13 min · 3,670 lines · $3.42
Epics, features, stories mean 86%, builds 85% to 88%

Build 1
85%
26 of 26 checks graded
18 min · 4,262 lines · $4.34

Build 2
85%
26 of 26 checks graded
19 min · 3,782 lines · $4.87

Build 3
87%
26 of 26 checks graded
14 min · 3,465 lines · $3.88

Build 4
85%
26 of 26 checks graded
15 min · 3,719 lines · $4.22

Build 5
88%
26 of 26 checks graded
16 min · 3,963 lines · $2.50
Sprint mean 67%, builds 52% to 92%

Build 1
52%
26 of 26 checks graded
12 min · 2,629 lines · $3.07

Build 2
52%
26 of 26 checks graded
10 min · 2,511 lines · $2.29

Build 3
92%
26 of 26 checks graded
19 min · 5,513 lines · $5.22

Build 4
85%
26 of 26 checks graded
19 min · 4,493 lines · $4.57

Build 5
52%
26 of 26 checks graded
9 min · 2,349 lines · $2.59