• There are 85 files with 35,452 lines of code.
• 9 very long files (15,553 lines of code)
• 12 long files (8,658 lines of code)
• 18 medium size files (6,783 lines of code)
• 22 small files (3,273 lines of code)
• 24 very small files (1,185 lines of code)

File	# lines	≈ tokens to read	# changes (1y)	≈ lines edited per change	lines read per line changed
Get_started_LiveAPI_tools.ipynb in quickstarts	2800	~15k–50k	11	~300	~10
Get_Started_Nano_Banana.ipynb in quickstarts	1534	~10k–30k	15	~500	~3
Video_understanding.ipynb in quickstarts	1249	~7k–20k	16	~100	~10
Get_started_thinking.ipynb in quickstarts	1247	~7k–20k	15	~500	~3
Get_started_OpenAI_Compatibility.ipynb in quickstarts	1509	~10k–30k	12	~200	~7
File_Search.ipynb in quickstarts	1131	~7k–20k	14	~150	~7
Models_REST.ipynb in quickstarts/rest	3106	~20k–50k	5	~200	~15
Function_calling.ipynb in quickstarts	1241	~7k–20k	12	~150	~7
Grounding.ipynb in quickstarts	919	~5k–15k	16	~150	~7
Get_started_managed_agents.ipynb in quickstarts	1736	~10k–30k	8	~200	~7
File_API.ipynb in quickstarts	789	~5k–15k	16	~70	~15
Spatial_understanding.ipynb in quickstarts	898	~5k–15k	14	~70	~10
Audio_REST.ipynb in quickstarts/rest	704	~3k–10k	14	~50	~15
Function_calling_REST.ipynb in quickstarts/rest	771	~5k–15k	12	~20	~30
JSON_mode.ipynb in quickstarts	595	~3k–10k	15	~30	~15
Caching.ipynb in quickstarts	679	~3k–10k	13	~70	~10
Counting_Tokens.ipynb in quickstarts	615	~3k–10k	13	~70	~10
Audio.ipynb in quickstarts	551	~3k–10k	14	~50	~10
Inference_tiers.ipynb in quickstarts	745	~5k–15k	9	~150	~5
Enum.ipynb in quickstarts	466	~3k–7k	14	~30	~15

Tokens are lines × 7 to 14 (analysis.fileReadsForChanges). Lines edited per change are the file's lines added and deleted divided by its commits over the whole history. Agents read about 2,000 lines at a time: when a 17,000-line file was split into modules, the tokens an agent read for the same change fell by 83% (The Economic Benefit of Refactoring, 2026).

There are 58 files with lines longer than 120 characters. In total, there are 1443 long lines.

File	# lines	# units	# long lines
Get_started_LiveAPI_tools.ipynb in quickstarts	2800	-	155
Video_understanding.ipynb in quickstarts	1249	-	133
Get_started_managed_agents.ipynb in quickstarts	1736	-	97
Get_started_OpenAI_Compatibility.ipynb in quickstarts	1509	-	79
Inference_tiers.ipynb in quickstarts	745	-	72
File_Search.ipynb in quickstarts	1131	-	68
Function_calling.ipynb in quickstarts	1241	-	61
Spatial_understanding.ipynb in quickstarts	898	-	49
File_API.ipynb in quickstarts	789	-	49
Grounding.ipynb in quickstarts	919	-	42
Get_Started_Nano_Banana.ipynb in quickstarts	1534	-	37
Get_started_thinking.ipynb in quickstarts	1247	-	35
Audio_REST.ipynb in quickstarts/rest	704	-	34
Authentication.ipynb in quickstarts	423	-	33
Get_started_Deep_Research.ipynb in quickstarts	824	-	32
Template.ipynb in quickstarts	344	-	32
Authentication_with_OAuth.ipynb in quickstarts	468	-	29
System_instructions.ipynb in quickstarts	463	-	29
Caching.ipynb in quickstarts	679	-	27
Audio.ipynb in quickstarts	551	-	26
Counting_Tokens.ipynb in quickstarts	615	-	26
JSON_mode.ipynb in quickstarts	595	-	23
Streaming.ipynb in quickstarts	313	-	22
Function_calling_REST.ipynb in quickstarts/rest	771	-	19
Get_started_interactions_api.ipynb in quickstarts	104	-	18
Safety_REST.ipynb in quickstarts/rest	493	-	18
Function_calling_config_REST.ipynb in quickstarts/rest	378	-	18
PDF_Files.ipynb in quickstarts	408	-	17
Webhooks.ipynb in quickstarts	568	-	17
Enum.ipynb in quickstarts	466	-	14
Image_out.js in quickstarts-js	496	1	12
Embeddings_REST.ipynb in quickstarts/rest	351	-	12
llm_judge.py in tools/nb_tester	162	2	10
JSON_mode_REST.ipynb in quickstarts/rest	191	-	10
Safety.ipynb in quickstarts	161	-	9
Models.ipynb in quickstarts	288	-	8
System_instructions_REST.ipynb in quickstarts/rest	248	-	7
Streaming_REST.ipynb in quickstarts/rest	139	-	7
Models_REST.ipynb in quickstarts/rest	3106	-	7
grounded_verifier.py in tools/nb_tester	140	2	6
Get_started_LyriaRealTime.py in quickstarts	154	-	6
security_scanner.py in tools/nb_tester	190	4	5
Get_started_LiveTranslate.py in quickstarts	187	1	5
ai_security_auditor.py in tools/nb_tester	145	2	3
Counting_Tokens.js in quickstarts-js	97	-	3
config.py in tools	122	1	2
model_override.py in tools/nb_tester	113	6	2
reporter.py in tools/nb_tester	133	5	2
gemini.py in tools/nblint/rules	114	5	2
model_selector.py in tools/nblint/rules	54	1	2

• File size measurements show the distribution of size of files.
• Files are classified in four categories based on their size (lines of code): 1-100 (very small files), 101-200 (small files), 201-500 (medium size files), 501-1000 (long files), 1001+(very long files).
• It is a good practice to keep files small. Long files may become "bloaters", code that have increased to such gargantuan proportions that they are hard to work with.