openai/build-hours
Unit Complexity*
The distribution of complexity of units (measured with McCabe index).
Conditional Complexity Overall
  • There are 1,842 units with 33,801 lines of code in units (33.8% of code).
    • 6 very complex units (3,008 lines of code)
    • 10 complex units (1,593 lines of code)
    • 81 medium complex units (4,604 lines of code)
    • 175 simple units (5,778 lines of code)
    • 1,570 very simple units (18,818 lines of code)
8% | 4% | 13% | 17% | 55%
Legend:
51+
26-50
11-25
6-10
1-5
Alternative Visuals
Conditional Complexity per Extension
51+
26-50
11-25
6-10
1-5
tsx11% | 9% | 5% | 11% | 61%
ts10% | 0% | 16% | 20% | 52%
py0% | 2% | 21% | 20% | 55%
js0% | 12% | 28% | 23% | 35%
Conditional Complexity per Logical Component
primary logical decomposition
51+
26-50
11-25
6-10
1-5
25-realtime-219% | 3% | 17% | 13% | 46%
16-built-in-tools35% | 0% | 0% | 17% | 46%
27-valuemaxxing14% | 0% | 15% | 21% | 48%
21-agentic-memory0% | 7% | 6% | 10% | 75%
24-api-codex0% | 18% | 28% | 29% | 24%
13-image-gen0% | 47% | 0% | 30% | 21%
22-chatgptapp-pingpong0% | 86% | 0% | 0% | 13%
18-responses-api0% | 0% | 24% | 18% | 57%
05-4o_fine_tuning0% | 0% | 51% | 9% | 39%
07-reasoning0% | 0% | 53% | 10% | 35%
09-realtime0% | 0% | 17% | 40% | 42%
14-voice-agents0% | 0% | 9% | 24% | 66%
20-agent-rft0% | 0% | 18% | 23% | 58%
15-reinforcement-fine-tuning0% | 0% | 19% | 24% | 55%
11-recommendation0% | 0% | 13% | 28% | 57%
12-agentic-tool-calling0% | 0% | 13% | 4% | 81%
02-assistants0% | 0% | 10% | 37% | 52%
26-agents-sdk0% | 0% | 0% | 13% | 86%
04-mmrag_tooluse0% | 0% | 0% | 28% | 71%
06-structured_outputs0% | 0% | 0% | 9% | 90%
17-gpt-50% | 0% | 0% | 33% | 66%
03-4o_mini_fine_tuning0% | 0% | 0% | 0% | 100%
08-evals0% | 0% | 0% | 0% | 100%
Most Complex Units
Top 50 most complex units
Unit# linesMcCabe index# params
661 112 0
339 87 4
334 87 4
764 70 0
424 55 0
486 55 0
277 49 7
204 42 0
63 35 1
77 34 1
173 30 6
29 29 1
141 29 2
298 28 0
162 28 0
169 27 0
62 25 1
52 24 1
39 24 1
106 24 0
21 23 1
28 23 1
40 23 1
33 20 1
115 20 2
86 20 2
33 19 1
155 19 8
87 19 1
52 19 1
56 19 1
70 19 5
68 19 0
110 18 3
37 18 1
20 18 1
62 18 2
34 17 1
105 17 1
214 17 13
57 17 0
37 17 1
21 16 1
209 16 7
44 16 4
60 16 0
34 15 1
95 15 8
42 15 1
24 15 1
About This Analysis
  • Conditional complexity (also called cyclomatic complexity) is a term used to measure the complexity of software. The term refers to the number of possible paths through a program function. A higher value ofter means higher maintenance and testing costs (infosecinstitute.com).
  • Conditional complexity is calculated by counting all conditions in the program that can affect the execution path (e.g. if statement, loops, switches, and/or operators, try and catch blocks...).
  • Conditional complexity is measured at the unit level (methods, functions...).
  • Units are classified in four categories based on the measured McCabe index: 1-5 (simple units), 6-10 (medium complex units), 11-25 (complex units), 26+ (very complex units).
Learn more...