head-to-head

Google: Gemini 3.5 Flash Lite (batch) vs Anthropic: Claude Sonnet 5 (batch)

Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-08-13.

	Google: Gemini 3.5 Flash Lite (batch)	Anthropic: Claude Sonnet 5 (batch)
Vendor	google	anthropic
Quality Score	100	100
Benchmark Score	62.3	91.6
Input Price	$0.15/M	$1.00/M
Output Price	$1.25/M	$5.00/M
Context Window	1,048,576	1,000,000
Max Output	65,536	128,000
Tool Calling	✓	✓
Structured Output	✓	✓
Reasoning Mode	✓	✓
Vision	✓	✓
Audio	✓	-
Benchmark Scores
ai_index	61.8	91.2
ai_index_agentic	44.8	82.0
ai_index_coding	81.4	100.0

Who wins by task?

Task	Google: Gemini 3.5 Flash Lite (batch)	Anthropic: Claude Sonnet 5 (batch)
SQL Generation	161	172
Code Review	157	168
Code Completion	133	120
Code Refactoring	156	165
Bug Fixing	166	182
Unit Test Generation	146	155
Code Documentation	140	143
Regex Writing	131	136
CI/CD Pipelines	137	145
Frontend Component Design	139	147
Data Analysis	157	172
CSV / Spreadsheet Cleanup	149	154
ETL Scripting	147	155
JSON Extraction	145	140
Bulk Data Labeling	134	126
OCR / Document Parsing	143	147
Table Extraction from PDFs	143	147
Long-Document Summarization	153	160
Short-Form Summarization	129	123
Blog Post Writing	135	141

Scores reflect capability match + benchmark data + pricing for each task. Methodology →